{ "openapi": "3.1.0", "info": { "title": "withoutbg API", "description": "\n\n \"logo\"\n\n\nCloud-hosted background removal and alpha matte extraction at\n`https://api.withoutbg.com`. Billed per successful image. Auth via `X-API-Key`.\n\n**Machine-readable spec (recommended LLM context):** [/openapi.json](/openapi.json) ·\n[/llms.txt](/llms.txt)\n\n## Shared rules\n\n| Rule | Value |\n|------|-------|\n| Base URL | `https://api.withoutbg.com` |\n| Auth header | `X-API-Key: YOUR_API_KEY` |\n| Rate limit (product) | 30 requests/minute per API key (gateway may return `429`) |\n| Max input size | 20.0 MB: binary: HTTP `413`; base64: HTTP `422` |\n| Unsupported format | Binary: HTTP `415`; base64: HTTP `422` |\n| Prep downscale | Longest side > 5000 px or > 25 MP → auto-downscale (not rejected) |\n| Paid call cost | 1 credit on HTTP 200 (credits endpoint is free) |\n| Inference size | Max ~1024 px on the longer side; matte/cutout mapped back to full-res RGB. Edge detail is capped at inference resolution. |\n\n**Binary vs Base64:** Binary (`multipart/form-data`) fits servers, CLIs, and file I/O.\nBase64 JSON fits browsers, serverless, and JSON-only pipelines. Send raw Base64\n(no `data:` URI prefix).\n\n**Cutout vs Alpha matte:** Cutout returns a ready RGBA PNG. Alpha matte returns a grayscale\nmask so you can keep a full-res original local, apply the matte yourself, and reuse one mask\nfor multiple composites.\n\n**Cars:** Prefer `/v1.0/car-*` for vehicle photos. Every car route crops to the\nbiggest detected vehicle before segmentation, then maps the result back to the\nfull frame. Cutout routes (`/v1.0/car-image-without-background` and `-base64`)\nrun window reconstruction and bake a cast shadow by default\n(`reconstruct_windows=true`, `shadow_type=ai`); they do not return `view`.\n`/v1.0/car-layers-base64` returns uncomposited cutout + shadow layers plus\n`view` (azimuth/elevation). `shadow_type` selects the renderer: `ai` (default,\nlearned model) or `none`. Light direction is inferred from the photo, so no\nshadow type takes light parameters.\n`view.azimuth_deg` is orbit around the car (`[0, 360)`; `0` = front, increasing\ncounterclockwise from above); `elevation_deg` is the look angle above the\nhorizon (`0` = level / mid-body; positive = camera above the car).\nNo image-type sidecar on car endpoints.\n\n**Supported formats:** JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF\n\n**Human docs:** [Pro Model API](https://withoutbg.com/docs/pro-model)\n", "contact": { "name": "withoutbg - contact", "url": "https://withoutbg.com/contact" }, "version": "1.5.0" }, "servers": [ { "url": "/", "description": "Current host" }, { "url": "https://api.withoutbg.com", "description": "Production" } ], "paths": { "/v1.0/alpha-channel": { "post": { "tags": [ "Alpha Matte" ], "summary": "Get alpha matte (PNG)", "description": "Return a grayscale alpha matte as a PNG.\n\n**Cost:** 1 credit on HTTP 200.\n\nPrefer this over the cutout endpoint when you need a full-resolution master: keep the original locally, optionally send a ~1024 px proxy, then apply the matte yourself. Edge detail is capped at model inference resolution (~1024 long side).\n\nBinary sibling of `/v1.0/alpha-channel-base64`. Response metadata is returned in `X-Input-Dimensions`, `X-Processed-Dimensions`, `X-Was-Downscaled`, and optional `X-Image-Type-*` headers.", "operationId": "alpha_channel_v1_0_alpha_channel_post", "requestBody": { "content": { "multipart/form-data": { "schema": { "$ref": "#/components/schemas/AlphaChannelRequest", "properties": { "file": { "type": "string", "format": "binary", "description": "Image file (JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF). Maximum size: 20.0 MB." }, "include_image_type": { "type": "boolean", "description": "When true (default), include photo vs graphic probabilities in X-Image-Type-* headers.", "default": true } }, "type": "object", "required": [ "file" ] } } }, "required": true }, "responses": { "200": { "description": "Success: PNG image body", "headers": { "X-Input-Dimensions": { "description": "Decoded input size as `width,height` (pixels).", "schema": { "type": "string", "example": "5568,3712" } }, "X-Processed-Dimensions": { "description": "Size after server prep downscale (longest side / megapixel caps) as `width,height`. Not the model inference resolution (~1024 long side).", "schema": { "type": "string", "example": "5000,3333" } }, "X-Was-Downscaled": { "description": "Whether the input was downscaled during server prep (`true`/`false`).", "schema": { "type": "string", "enum": [ "true", "false" ], "example": "true" } }, "X-Image-Type-Photo": { "description": "Optional photo-class probability from image-type classifier.", "schema": { "type": "string", "example": "0.87" } }, "X-Image-Type-Graphic": { "description": "Optional graphic-class probability from image-type classifier.", "schema": { "type": "string", "example": "0.13" } } }, "content": { "image/png": {} } }, "401": { "description": "Invalid API Key", "content": { "application/json": { "example": { "detail": "Invalid API Key" } } } }, "402": { "description": "Insufficient credit. Please top up API credits.", "content": { "application/json": { "example": { "detail": "Insufficient credit. Please top up API credits." } } } }, "403": { "description": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated.", "content": { "application/json": { "example": { "detail": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated." } } } }, "413": { "description": "File size too large. Maximum file size is 20.0 MB.", "content": { "application/json": { "example": { "detail": "File size too large. Maximum file size is 20.0 MB" } } } }, "415": { "description": "Unsupported Media Type. Supported formats are: JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF.", "content": { "application/json": { "example": { "detail": "Unsupported Media Type. Supported formats are: JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF." } } } }, "422": { "description": "Validation Error", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/HTTPValidationError" } } } }, "429": { "description": "Rate limit exceeded (product policy: 30 requests/minute per API key). May be returned by the API gateway. Retry with exponential backoff.", "content": { "application/json": { "example": { "error": "Too Many Requests", "status": 429, "message": "Rate limit exceeded. Please try again later" } } } }, "500": { "description": "Internal Server Error", "content": { "application/json": { "example": { "detail": "Internal Server Error. Please contact support: contact@withoutbg.com" } } } } }, "security": [ { "APIKeyHeader": [] } ] } }, "/v1.0/image-without-background": { "post": { "tags": [ "Background Removal" ], "summary": "Remove background (PNG)", "description": "Remove the background and return an RGBA PNG cutout.\n\n**Cost:** 1 credit on HTTP 200.\n\nBest for servers, CLIs, and batch pipelines that want raw PNG bytes (no JSON/base64 overhead). For a full-resolution compositing workflow, prefer the alpha-matte endpoints.\n\nBinary sibling of `/v1.0/image-without-background-base64`. Response metadata is returned in `X-Input-Dimensions`, `X-Processed-Dimensions`, `X-Was-Downscaled`, and optional `X-Image-Type-*` headers.", "operationId": "with_transparent_background_v1_0_image_without_background_post", "requestBody": { "content": { "multipart/form-data": { "schema": { "$ref": "#/components/schemas/ImageWithoutBackgroundRequest", "properties": { "file": { "type": "string", "format": "binary", "description": "Image file (JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF). Maximum size: 20.0 MB." }, "include_image_type": { "type": "boolean", "description": "When true (default), include photo vs graphic probabilities in X-Image-Type-* headers.", "default": true } }, "type": "object", "required": [ "file" ] } } }, "required": true }, "responses": { "200": { "description": "Success: PNG image body", "headers": { "X-Input-Dimensions": { "description": "Decoded input size as `width,height` (pixels).", "schema": { "type": "string", "example": "5568,3712" } }, "X-Processed-Dimensions": { "description": "Size after server prep downscale (longest side / megapixel caps) as `width,height`. Not the model inference resolution (~1024 long side).", "schema": { "type": "string", "example": "5000,3333" } }, "X-Was-Downscaled": { "description": "Whether the input was downscaled during server prep (`true`/`false`).", "schema": { "type": "string", "enum": [ "true", "false" ], "example": "true" } }, "X-Image-Type-Photo": { "description": "Optional photo-class probability from image-type classifier.", "schema": { "type": "string", "example": "0.87" } }, "X-Image-Type-Graphic": { "description": "Optional graphic-class probability from image-type classifier.", "schema": { "type": "string", "example": "0.13" } } }, "content": { "image/png": {} } }, "401": { "description": "Invalid API Key", "content": { "application/json": { "example": { "detail": "Invalid API Key" } } } }, "402": { "description": "Insufficient credit. Please top up API credits.", "content": { "application/json": { "example": { "detail": "Insufficient credit. Please top up API credits." } } } }, "403": { "description": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated.", "content": { "application/json": { "example": { "detail": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated." } } } }, "413": { "description": "File size too large. Maximum file size is 20.0 MB.", "content": { "application/json": { "example": { "detail": "File size too large. Maximum file size is 20.0 MB" } } } }, "415": { "description": "Unsupported Media Type. Supported formats are: JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF.", "content": { "application/json": { "example": { "detail": "Unsupported Media Type. Supported formats are: JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF." } } } }, "422": { "description": "Validation Error", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/HTTPValidationError" } } } }, "429": { "description": "Rate limit exceeded (product policy: 30 requests/minute per API key). May be returned by the API gateway. Retry with exponential backoff.", "content": { "application/json": { "example": { "error": "Too Many Requests", "status": 429, "message": "Rate limit exceeded. Please try again later" } } } }, "500": { "description": "Internal Server Error", "content": { "application/json": { "example": { "detail": "Internal Server Error. Please contact support: contact@withoutbg.com" } } } } }, "security": [ { "APIKeyHeader": [] } ] } }, "/v1.0/image-without-background-base64": { "post": { "tags": [ "Background Removal" ], "summary": "Remove background (Base64 JSON)", "description": "Remove the background and return a Base64 RGBA PNG in JSON.\n\n**Cost:** 1 credit on HTTP 200.\n\nBest for browsers, serverless, and JSON-only pipelines. Send a raw Base64 string (no `data:` URI prefix). Binary sibling: `/v1.0/image-without-background`.\n\nFor full-resolution compositing, prefer the alpha-matte endpoints.", "operationId": "image_without_background_base64_v1_0_image_without_background_base64_post", "requestBody": { "content": { "application/json": { "schema": { "$ref": "#/components/schemas/ImageBase64Request" } } }, "required": true }, "responses": { "200": { "description": "Success", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/ImageWithoutBackgroundResponse" } } } }, "401": { "description": "Invalid API Key", "content": { "application/json": { "example": { "detail": "Invalid API Key" } } } }, "402": { "description": "Insufficient credit. Please top up API credits.", "content": { "application/json": { "example": { "detail": "Insufficient credit. Please top up API credits." } } } }, "403": { "description": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated.", "content": { "application/json": { "example": { "detail": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated." } } } }, "422": { "description": "Validation error. Includes invalid Base64, non-image payloads, unsupported formats, and decoded size over 20.0 MB.", "content": { "application/json": { "example": { "detail": [ { "loc": [ "body", "image_base64" ], "msg": "Image size exceeds maximum allowed size of 20.0 MB", "type": "value_error" } ] } } } }, "429": { "description": "Rate limit exceeded (product policy: 30 requests/minute per API key). May be returned by the API gateway. Retry with exponential backoff.", "content": { "application/json": { "example": { "error": "Too Many Requests", "status": 429, "message": "Rate limit exceeded. Please try again later" } } } }, "500": { "description": "Internal Server Error", "content": { "application/json": { "example": { "detail": "Internal Server Error. Please contact support: contact@withoutbg.com" } } } } }, "security": [ { "APIKeyHeader": [] } ] } }, "/v1.0/alpha-channel-base64": { "post": { "tags": [ "Alpha Matte" ], "summary": "Get alpha matte (Base64 JSON)", "description": "Return a grayscale alpha matte as Base64 PNG in JSON.\n\n**Cost:** 1 credit on HTTP 200.\n\nRecommended when you keep a full-res original locally, send a ~1024 px proxy, then upscale/apply the matte yourself. Binary sibling: `/v1.0/alpha-channel`.", "operationId": "alpha_channel_base64_v1_0_alpha_channel_base64_post", "requestBody": { "content": { "application/json": { "schema": { "$ref": "#/components/schemas/ImageBase64Request" } } }, "required": true }, "responses": { "200": { "description": "Success", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/AlphaChannelResponse" } } } }, "401": { "description": "Invalid API Key", "content": { "application/json": { "example": { "detail": "Invalid API Key" } } } }, "402": { "description": "Insufficient credit. Please top up API credits.", "content": { "application/json": { "example": { "detail": "Insufficient credit. Please top up API credits." } } } }, "403": { "description": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated.", "content": { "application/json": { "example": { "detail": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated." } } } }, "422": { "description": "Validation error. Includes invalid Base64, non-image payloads, unsupported formats, and decoded size over 20.0 MB.", "content": { "application/json": { "example": { "detail": [ { "loc": [ "body", "image_base64" ], "msg": "Image size exceeds maximum allowed size of 20.0 MB", "type": "value_error" } ] } } } }, "429": { "description": "Rate limit exceeded (product policy: 30 requests/minute per API key). May be returned by the API gateway. Retry with exponential backoff.", "content": { "application/json": { "example": { "error": "Too Many Requests", "status": 429, "message": "Rate limit exceeded. Please try again later" } } } }, "500": { "description": "Internal Server Error", "content": { "application/json": { "example": { "detail": "Internal Server Error. Please contact support: contact@withoutbg.com" } } } } }, "security": [ { "APIKeyHeader": [] } ] } }, "/v1.0/car-image-without-background": { "post": { "tags": [ "Car Background Removal" ], "summary": "Remove car background (PNG)", "description": "Remove background from a car photo and return an RGBA PNG.\n\n**Cost:** 1 credit on HTTP 200.\n\nUses the dedicated car pipeline: the biggest vehicle is detected and cropped before segmentation, then results are mapped back to the full frame. Window reconstruction and a baked cast shadow run by default (`reconstruct_windows=true`, `shadow_type=ai`). Set `shadow_type=none` for a cutout with no baked shadow. No camera `view` sidecar (use `/v1.0/car-layers-base64` for compositing layers plus that sidecar). Image-type classification is not offered (car photos are treated as photographic).\n\nBinary sibling of `/v1.0/car-image-without-background-base64`. Response metadata is returned in `X-Input-Dimensions`, `X-Processed-Dimensions`, `X-Was-Downscaled`, and `X-Shadow-Type`.", "operationId": "car_image_without_background_v1_0_car_image_without_background_post", "requestBody": { "content": { "multipart/form-data": { "schema": { "$ref": "#/components/schemas/CarImageWithoutBackgroundRequest" } } }, "required": true }, "responses": { "200": { "description": "Success: PNG image body", "headers": { "X-Input-Dimensions": { "description": "Decoded input size as `width,height` (pixels).", "schema": { "type": "string", "example": "5568,3712" } }, "X-Processed-Dimensions": { "description": "Size after server prep downscale (longest side / megapixel caps) as `width,height`. Not the model inference resolution (~1024 long side).", "schema": { "type": "string", "example": "5000,3333" } }, "X-Was-Downscaled": { "description": "Whether the input was downscaled during server prep (`true`/`false`).", "schema": { "type": "string", "enum": [ "true", "false" ], "example": "true" } }, "X-Shadow-Type": { "description": "Shadow renderer that produced the baked cast shadow ('ai', or 'none' when no shadow was requested).", "schema": { "type": "string", "example": "ai" } } }, "content": { "image/png": {} } }, "401": { "description": "Invalid API Key", "content": { "application/json": { "example": { "detail": "Invalid API Key" } } } }, "402": { "description": "Insufficient credit. Please top up API credits.", "content": { "application/json": { "example": { "detail": "Insufficient credit. Please top up API credits." } } } }, "403": { "description": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated.", "content": { "application/json": { "example": { "detail": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated." } } } }, "413": { "description": "File size too large. Maximum file size is 20.0 MB.", "content": { "application/json": { "example": { "detail": "File size too large. Maximum file size is 20.0 MB" } } } }, "415": { "description": "Unsupported Media Type. Supported formats are: JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF.", "content": { "application/json": { "example": { "detail": "Unsupported Media Type. Supported formats are: JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF." } } } }, "422": { "description": "Validation Error", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/HTTPValidationError" } } } }, "429": { "description": "Rate limit exceeded (product policy: 30 requests/minute per API key). May be returned by the API gateway. Retry with exponential backoff.", "content": { "application/json": { "example": { "error": "Too Many Requests", "status": 429, "message": "Rate limit exceeded. Please try again later" } } } }, "500": { "description": "Internal Server Error", "content": { "application/json": { "example": { "detail": "Internal Server Error. Please contact support: contact@withoutbg.com" } } } } }, "security": [ { "APIKeyHeader": [] } ] } }, "/v1.0/car-image-without-background-base64": { "post": { "tags": [ "Car Background Removal" ], "summary": "Remove car background (Base64 JSON)", "description": "Remove background from a car photo; return Base64 RGBA PNG in JSON.\n\n**Cost:** 1 credit on HTTP 200.\n\nUses the dedicated car pipeline: the biggest vehicle is detected and cropped before segmentation, then results are mapped back to the full frame. Window reconstruction and a baked cast shadow run by default (`reconstruct_windows=true`, `shadow_type=ai`). Set `shadow_type=none` to skip the shadow. No camera `view` sidecar (use `/v1.0/car-layers-base64` for compositing layers plus that sidecar). No image-type sidecar (car photos are photographic). Binary sibling: `/v1.0/car-image-without-background`.", "operationId": "car_image_without_background_base64_v1_0_car_image_without_background_base64_post", "requestBody": { "content": { "application/json": { "schema": { "$ref": "#/components/schemas/CarImageBase64Request" } } }, "required": true }, "responses": { "200": { "description": "Success", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/CarImageWithoutBackgroundResponse" } } } }, "401": { "description": "Invalid API Key", "content": { "application/json": { "example": { "detail": "Invalid API Key" } } } }, "402": { "description": "Insufficient credit. Please top up API credits.", "content": { "application/json": { "example": { "detail": "Insufficient credit. Please top up API credits." } } } }, "403": { "description": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated.", "content": { "application/json": { "example": { "detail": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated." } } } }, "422": { "description": "Validation error. Includes invalid Base64, non-image payloads, unsupported formats, and decoded size over 20.0 MB.", "content": { "application/json": { "example": { "detail": [ { "loc": [ "body", "image_base64" ], "msg": "Image size exceeds maximum allowed size of 20.0 MB", "type": "value_error" } ] } } } }, "429": { "description": "Rate limit exceeded (product policy: 30 requests/minute per API key). May be returned by the API gateway. Retry with exponential backoff.", "content": { "application/json": { "example": { "error": "Too Many Requests", "status": 429, "message": "Rate limit exceeded. Please try again later" } } } }, "500": { "description": "Internal Server Error", "content": { "application/json": { "example": { "detail": "Internal Server Error. Please contact support: contact@withoutbg.com" } } } } }, "security": [ { "APIKeyHeader": [] } ] } }, "/v1.0/car-layers-base64": { "post": { "tags": [ "Car Compositing" ], "summary": "Car compositing layers (Base64 JSON)", "description": "Remove background from a car photo and return the RGBA cutout and a soft cast-shadow mask as separate Base64 PNGs, plus a camera `view` sidecar.\n\n**Cost:** 1 credit on HTTP 200.\n\nThe cutout does **not** bake the shadow in; use `shadow_base64` to composite or edit on the client. Window reconstruction is on by default (`reconstruct_windows=true`); set it to `false` to skip. `shadow_type=none` omits `shadow_base64`. For a single RGBA with baked shadow, use `/v1.0/car-image-without-background-base64`. No binary sibling. No image-type sidecar.", "operationId": "car_layers_base64_v1_0_car_layers_base64_post", "requestBody": { "content": { "application/json": { "schema": { "$ref": "#/components/schemas/CarLayersBase64Request" } } }, "required": true }, "responses": { "200": { "description": "Success", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/CarLayersResponse" } } } }, "401": { "description": "Invalid API Key", "content": { "application/json": { "example": { "detail": "Invalid API Key" } } } }, "402": { "description": "Insufficient credit. Please top up API credits.", "content": { "application/json": { "example": { "detail": "Insufficient credit. Please top up API credits." } } } }, "403": { "description": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated.", "content": { "application/json": { "example": { "detail": "Credits Expired. Please top up API credits. If you have existing credits, they will be reactivated." } } } }, "422": { "description": "Validation error. Includes invalid Base64, non-image payloads, unsupported formats, and decoded size over 20.0 MB.", "content": { "application/json": { "example": { "detail": [ { "loc": [ "body", "image_base64" ], "msg": "Image size exceeds maximum allowed size of 20.0 MB", "type": "value_error" } ] } } } }, "429": { "description": "Rate limit exceeded (product policy: 30 requests/minute per API key). May be returned by the API gateway. Retry with exponential backoff.", "content": { "application/json": { "example": { "error": "Too Many Requests", "status": 429, "message": "Rate limit exceeded. Please try again later" } } } }, "500": { "description": "Internal Server Error", "content": { "application/json": { "example": { "detail": "Internal Server Error. Please contact support: contact@withoutbg.com" } } } } }, "security": [ { "APIKeyHeader": [] } ] } }, "/available-credit": { "get": { "tags": [ "API Credits" ], "summary": "Get available API credits", "description": "Return the current credit balance and expiration date for the API key.\n\n**Cost:** free (does not consume credits).\n\nUse for monitoring, dashboards, pre-flight checks, and recovering from 402 (insufficient credits) or 403 (expired credits) on paid endpoints.", "operationId": "get_your_available_credit_available_credit_get", "responses": { "200": { "description": "Success", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/CreditInfo" } } } }, "401": { "description": "Invalid API Key", "content": { "application/json": { "example": { "detail": "Invalid API Key" } } } }, "429": { "description": "Rate limit exceeded (product policy: 30 requests/minute per API key). May be returned by the API gateway. Retry with exponential backoff.", "content": { "application/json": { "example": { "error": "Too Many Requests", "status": 429, "message": "Rate limit exceeded. Please try again later" } } } }, "500": { "description": "Internal Server Error", "content": { "application/json": { "example": { "detail": "Internal Server Error. Please contact support: contact@withoutbg.com" } } } } }, "security": [ { "APIKeyHeader": [] } ] } } }, "components": { "schemas": { "AlphaChannelRequest": { "properties": { "file": { "type": "string", "format": "binary", "title": "File" }, "include_image_type": { "type": "boolean", "title": "Include Image Type", "description": "When true (default), include photo vs graphic probabilities in X-Image-Type-* headers. Set false to skip.", "default": true } }, "type": "object", "required": [ "file" ], "title": "AlphaChannelRequest" }, "AlphaChannelResponse": { "properties": { "input_dimensions": { "items": { "type": "integer" }, "type": "array", "title": "Input Dimensions", "description": "Decoded input size as `[width, height]` (pixels).", "examples": [ [ 5568, 3712 ] ] }, "processed_dimensions": { "items": { "type": "integer" }, "type": "array", "title": "Processed Dimensions", "description": "Size after server prep downscale (longest-side / megapixel caps) as `[width, height]`. Distinct from model inference resolution (~1024 long side).", "examples": [ [ 5000, 3333 ] ] }, "was_downscaled": { "type": "boolean", "title": "Was Downscaled", "description": "True when the input was downscaled during server prep.", "examples": [ true ] }, "image_type": { "anyOf": [ { "additionalProperties": { "type": "number" }, "type": "object" }, { "type": "null" } ], "title": "Image Type", "description": "Photo vs graphic class probabilities. Present when `include_image_type` is true (default); omitted when false. Never returned by car endpoints.", "examples": [ { "graphic": 0.13, "photo": 0.87 } ] }, "alpha_base64": { "type": "string", "title": "Alpha Base64", "description": "Grayscale alpha matte as Base64 PNG. Prepend `data:image/png;base64,` to display in a browser.", "examples": [ "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mP8z8BQDwAEhQGAhKmMIQAAAABJRU5ErkJggg==" ] } }, "type": "object", "required": [ "input_dimensions", "processed_dimensions", "was_downscaled", "alpha_base64" ], "title": "AlphaChannelResponse", "description": "Base64 alpha channel matte." }, "CarImageBase64Request": { "properties": { "image_base64": { "type": "string", "title": "Image Base64", "description": "Input image as a raw Base64 string (no `data:` URI prefix). Supported formats: JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF. Maximum decoded size: 20.0 MB. Invalid Base64, non-image data, unsupported formats, or oversize payloads fail with HTTP 422.", "examples": [ "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==" ] }, "shadow_type": { "anyOf": [ { "$ref": "#/components/schemas/ShadowType" }, { "type": "null" } ], "description": "Which renderer produces the cast shadow. Defaults to `ai`, the learned car-shadow model. `none` skips the shadow entirely. `geometric` (proxy fit) and `auto` (geometric, falling back to `ai`) are declared but not yet available and are rejected with 422; light direction is inferred from the photo, so no shadow type takes light parameters." }, "include_shadow": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "title": "Include Shadow", "description": "Deprecated: use `shadow_type` instead (true maps to `ai`, false to `none`). When omitted, `shadow_type` decides. Setting both to contradictory values is rejected.", "deprecated": true }, "reconstruct_windows": { "type": "boolean", "title": "Reconstruct Windows", "description": "When true (default), run car window reconstruction: opens glass interiors while keeping the car outline. Cutout RGB stays the original photo unless a shadow is baked in. Set false to skip and use the base silhouette alpha only.", "default": true } }, "type": "object", "required": [ "image_base64" ], "title": "CarImageBase64Request", "description": "Car remove-bg request (always photographic; no image-type sidecar)." }, "CarImageWithoutBackgroundRequest": { "properties": { "file": { "type": "string", "format": "binary", "title": "File", "description": "Car image file (JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF). Maximum size: 20.0 MB." }, "shadow_type": { "anyOf": [ { "$ref": "#/components/schemas/ShadowType" }, { "type": "null" } ], "description": "Which renderer produces the cast shadow. Defaults to `ai`, the learned car-shadow model. `none` skips the shadow entirely. `geometric` (proxy fit) and `auto` (geometric, falling back to `ai`) are declared but not yet available and are rejected with 422; light direction is inferred from the photo, so no shadow type takes light parameters." }, "include_shadow": { "anyOf": [ { "type": "boolean" }, { "type": "null" } ], "title": "Include Shadow", "description": "Deprecated: use `shadow_type` instead (true maps to `ai`, false to `none`). When omitted, `shadow_type` decides.", "deprecated": true }, "reconstruct_windows": { "type": "boolean", "title": "Reconstruct Windows", "description": "When true (default), run car window reconstruction: opens glass interiors while keeping the car outline. Cutout RGB stays the original photo unless a shadow is baked in. Set false to skip and use the base silhouette alpha only.", "default": true } }, "type": "object", "required": [ "file" ], "title": "CarImageWithoutBackgroundRequest" }, "CarImageWithoutBackgroundResponse": { "properties": { "input_dimensions": { "items": { "type": "integer" }, "type": "array", "title": "Input Dimensions", "description": "Decoded input size as `[width, height]` (pixels).", "examples": [ [ 5568, 3712 ] ] }, "processed_dimensions": { "items": { "type": "integer" }, "type": "array", "title": "Processed Dimensions", "description": "Size after server prep downscale (longest-side / megapixel caps) as `[width, height]`. Distinct from model inference resolution (~1024 long side).", "examples": [ [ 5000, 3333 ] ] }, "was_downscaled": { "type": "boolean", "title": "Was Downscaled", "description": "True when the input was downscaled during server prep.", "examples": [ true ] }, "shadow_type": { "$ref": "#/components/schemas/ShadowType", "description": "The shadow renderer that actually ran for this request.", "examples": [ "ai" ] }, "img_without_background_base64": { "type": "string", "title": "Img Without Background Base64", "description": "Car cutout as Base64 PNG (RGBA) with transparent background. Unless `shadow_type` was `none`, the soft cast shadow is already baked into this image. Prepend `data:image/png;base64,` to display in a browser.", "examples": [ "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==" ] } }, "type": "object", "required": [ "input_dimensions", "processed_dimensions", "was_downscaled", "shadow_type", "img_without_background_base64" ], "title": "CarImageWithoutBackgroundResponse", "description": "Car Base64 cutout (no image-type or view sidecars)." }, "CarLayersBase64Request": { "properties": { "image_base64": { "type": "string", "title": "Image Base64", "description": "Input image as a raw Base64 string (no `data:` URI prefix). Supported formats: JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF. Maximum decoded size: 20.0 MB. Invalid Base64, non-image data, unsupported formats, or oversize payloads fail with HTTP 422.", "examples": [ "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==" ] }, "shadow_type": { "$ref": "#/components/schemas/ShadowType", "description": "Which renderer produces the cast shadow. Defaults to `ai`, the learned car-shadow model. `none` skips the shadow entirely. `geometric` (proxy fit) and `auto` (geometric, falling back to `ai`) are declared but not yet available and are rejected with 422; light direction is inferred from the photo, so no shadow type takes light parameters.", "default": "ai" }, "reconstruct_windows": { "type": "boolean", "title": "Reconstruct Windows", "description": "When true (default), run car window reconstruction on the cutout: opens glass interiors while keeping the car outline. Shadow is predicted from the BiRefNet silhouette regardless. Set false to skip window reconstruction.", "default": true } }, "type": "object", "required": [ "image_base64" ], "title": "CarLayersBase64Request", "description": "Car compositing layers: uncomposited cutout + soft shadow." }, "CarLayersResponse": { "properties": { "input_dimensions": { "items": { "type": "integer" }, "type": "array", "title": "Input Dimensions", "description": "Decoded input size as `[width, height]` (pixels).", "examples": [ [ 5568, 3712 ] ] }, "processed_dimensions": { "items": { "type": "integer" }, "type": "array", "title": "Processed Dimensions", "description": "Size after server prep downscale (longest-side / megapixel caps) as `[width, height]`. Distinct from model inference resolution (~1024 long side).", "examples": [ [ 5000, 3333 ] ] }, "was_downscaled": { "type": "boolean", "title": "Was Downscaled", "description": "True when the input was downscaled during server prep.", "examples": [ true ] }, "view": { "anyOf": [ { "$ref": "#/components/schemas/CarViewMetadata" }, { "type": "null" } ], "description": "Estimated camera orbit around the car. Omitted if the orientation head is not loaded. Azimuth 0° = front of the car, increasing counterclockwise from above. Elevation 0° = level look-at (mid-body height); positive = camera above the car." }, "shadow_type": { "$ref": "#/components/schemas/ShadowType", "description": "The shadow renderer that actually ran for this request.", "examples": [ "ai" ] }, "img_without_background_base64": { "type": "string", "title": "Img Without Background Base64", "description": "Car cutout as Base64 PNG (RGBA) with transparent background. Shadow is not baked in; use `shadow_base64` to composite yourself. Prepend `data:image/png;base64,` to display in a browser.", "examples": [ "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==" ] }, "shadow_base64": { "anyOf": [ { "type": "string" }, { "type": "null" } ], "title": "Shadow Base64", "description": "Soft cast-shadow mask as Base64 grayscale PNG (0 = no shadow, 255 = full shadow). Not composited into the cutout; for client-side editing/compositing. Null when `shadow_type` was `none`. Prepend `data:image/png;base64,` to display in a browser.", "examples": [ "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mP8z8BQDwAEhQGAhKmMIQAAAABJRU5ErkJggg==" ] } }, "type": "object", "required": [ "input_dimensions", "processed_dimensions", "was_downscaled", "shadow_type", "img_without_background_base64" ], "title": "CarLayersResponse", "description": "Base64 car RGBA cutout plus separate soft cast-shadow mask." }, "CarViewMetadata": { "properties": { "azimuth_deg": { "type": "number", "title": "Azimuth Deg", "description": "Camera orbit around the car, in degrees, range [0, 360). 0 looks at the front of the car. Increases counterclockwise from above (90 = camera on the car's left; nose points right in the photo).", "examples": [ 184.06 ] }, "elevation_deg": { "type": "number", "title": "Elevation Deg", "description": "Angle of the camera-to-car look direction above the horizon, in degrees. 0 is a level view (camera at look-at height, roughly mid-body, not ground). Positive means the camera is above the car.", "examples": [ 1.06 ] } }, "type": "object", "required": [ "azimuth_deg", "elevation_deg" ], "title": "CarViewMetadata", "description": "Estimated camera orbit around the car (not a full extrinsic / focal length).\n\nAzimuth is degrees in ``[0, 360)``: ``0`` looks at the **front** of the car\nand increases **counterclockwise from above** (``90`` = camera on the car's\nleft; the nose points right in the photo).\n\nElevation is the camera-to-car look angle above the horizon, in degrees:\n``0`` is a level view (camera at look-at height, roughly mid-body, not\nground). Positive means the camera is **above** the car." }, "CreditInfo": { "properties": { "credit": { "type": "integer", "title": "Credit", "description": "Credits available to spend right now.", "examples": [ 138 ] }, "expiration_date": { "type": "string", "title": "Expiration Date", "description": "When credits expire, as `YYYY-MM-DD HH:MM:SS`.", "examples": [ "2026-12-31 23:59:59" ] } }, "type": "object", "required": [ "credit", "expiration_date" ], "title": "CreditInfo" }, "HTTPValidationError": { "properties": { "detail": { "items": { "$ref": "#/components/schemas/ValidationError" }, "type": "array", "title": "Detail" } }, "type": "object", "title": "HTTPValidationError" }, "ImageBase64Request": { "properties": { "image_base64": { "type": "string", "title": "Image Base64", "description": "Input image as a raw Base64 string (no `data:` URI prefix). Supported formats: JPEG, PNG, WebP, AVIF, HEIC, TIFF, BMP, GIF. Maximum decoded size: 20.0 MB. Invalid Base64, non-image data, unsupported formats, or oversize payloads fail with HTTP 422.", "examples": [ "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==" ] }, "include_image_type": { "type": "boolean", "title": "Include Image Type", "description": "When true (default), include photo vs graphic probabilities in the response (`image_type` / `X-Image-Type-*`). Set false to skip.", "default": true } }, "type": "object", "required": [ "image_base64" ], "title": "ImageBase64Request", "description": "Base64 image model with optional image-type sidecar." }, "ImageWithoutBackgroundRequest": { "properties": { "file": { "type": "string", "format": "binary", "title": "File" }, "include_image_type": { "type": "boolean", "title": "Include Image Type", "description": "When true (default), include photo vs graphic probabilities in X-Image-Type-* headers. Set false to skip.", "default": true } }, "type": "object", "required": [ "file" ], "title": "ImageWithoutBackgroundRequest" }, "ImageWithoutBackgroundResponse": { "properties": { "input_dimensions": { "items": { "type": "integer" }, "type": "array", "title": "Input Dimensions", "description": "Decoded input size as `[width, height]` (pixels).", "examples": [ [ 5568, 3712 ] ] }, "processed_dimensions": { "items": { "type": "integer" }, "type": "array", "title": "Processed Dimensions", "description": "Size after server prep downscale (longest-side / megapixel caps) as `[width, height]`. Distinct from model inference resolution (~1024 long side).", "examples": [ [ 5000, 3333 ] ] }, "was_downscaled": { "type": "boolean", "title": "Was Downscaled", "description": "True when the input was downscaled during server prep.", "examples": [ true ] }, "image_type": { "anyOf": [ { "additionalProperties": { "type": "number" }, "type": "object" }, { "type": "null" } ], "title": "Image Type", "description": "Photo vs graphic class probabilities. Present when `include_image_type` is true (default); omitted when false. Never returned by car endpoints.", "examples": [ { "graphic": 0.13, "photo": 0.87 } ] }, "img_without_background_base64": { "type": "string", "title": "Img Without Background Base64", "description": "Result image as Base64 PNG (RGBA) with transparent background. Prepend `data:image/png;base64,` to display in a browser.", "examples": [ "iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAYAAAAfFcSJAAAADUlEQVR42mNk+M9QDwADhgGAWjR9awAAAABJRU5ErkJggg==" ] } }, "type": "object", "required": [ "input_dimensions", "processed_dimensions", "was_downscaled", "img_without_background_base64" ], "title": "ImageWithoutBackgroundResponse", "description": "Base64 image with transparent background." }, "ShadowType": { "type": "string", "enum": [ "none", "ai", "geometric", "auto" ], "title": "ShadowType", "description": "Which renderer produces the cast shadow." }, "ValidationError": { "properties": { "loc": { "items": { "anyOf": [ { "type": "string" }, { "type": "integer" } ] }, "type": "array", "title": "Location" }, "msg": { "type": "string", "title": "Message" }, "type": { "type": "string", "title": "Error Type" } }, "type": "object", "required": [ "loc", "msg", "type" ], "title": "ValidationError" } }, "securitySchemes": { "APIKeyHeader": { "type": "apiKey", "description": "API key for authentication. Pass as `X-API-Key` on every request.", "in": "header", "name": "X-API-Key" } } }, "tags": [ { "name": "Background Removal", "description": "Remove image background and return an RGBA PNG cutout (transparent background).", "externalDocs": { "description": "Detailed Documentation", "url": "https://withoutbg.com/docs/pro-model/background-removal-binary" } }, { "name": "Car Background Removal", "description": "Car-photo background removal (dedicated vehicle pipeline: YOLO vehicle crop, then BiRefNet). Window reconstruction and a baked cast shadow are on by default; set reconstruct_windows=false or shadow_type=none to skip. No image-type or view sidecars. For uncomposited cutout + shadow layers plus view, use POST /v1.0/car-layers-base64 (Car Compositing).", "externalDocs": { "description": "Detailed Documentation", "url": "https://withoutbg.com/docs/pro-model/car-background-removal-binary" } }, { "name": "Car Compositing", "description": "Car compositing layers: RGBA cutout and a separate soft shadow mask (not baked), plus view (camera azimuth/elevation). Window reconstruction is on by default. azimuth 0° = front of the car, increasing counterclockwise from above; elevation 0° = level look-at, positive = camera above the car.", "externalDocs": { "description": "Detailed Documentation", "url": "https://withoutbg.com/docs/pro-model/car-layers-base64" } }, { "name": "Alpha Matte", "description": "Generate a grayscale alpha matte (opacity mask) for compositing onto your own full-resolution RGB.", "externalDocs": { "description": "Detailed Documentation", "url": "https://withoutbg.com/docs/pro-model/alpha-matte-binary" } }, { "name": "API Credits", "description": "Returns your available credits and the expiration date.", "externalDocs": { "description": "Detailed Documentation", "url": "https://withoutbg.com/docs/pro-model/credits" } } ] }