BRIA RMBG vs BiRefNet General
37 examples, 37 with ground truth. Lower is better. Use the view controls on each row to switch between transparency, chroma key, and alpha matte on all outputs.
| Metric | BRIA RMBG | BiRefNet General |
|---|---|---|
| MGE (Edge Quality) | 0.06124 ✓ | 0.06476 |
| MAE (Overall Accuracy) | 0.04669 ✓ | 0.05694 |
| Connectivity | 0.04595 ✓ | 0.05636 |
- MGE
- 0.05965
- MAE
- 0.01411
- Conn.
- 0.01214
- MGE
- 0.05164✓
- MAE
- 0.01051✓
- Conn.
- 0.00860✓
- MGE
- 0.05241✓
- MAE
- 0.02103✓
- Conn.
- 0.02029✓
- MGE
- 0.06089
- MAE
- 0.03207
- Conn.
- 0.03135
- MGE
- 0.05665✓
- MAE
- 0.01132
- Conn.
- 0.01094✓
- MGE
- 0.06130
- MAE
- 0.01128✓
- Conn.
- 0.01096
- MGE
- 0.01510✓
- MAE
- 0.00660✓
- Conn.
- 0.00469✓
- MGE
- 0.01673
- MAE
- 0.00895
- Conn.
- 0.00713
- MGE
- 0.03261
- MAE
- 0.00697✓
- Conn.
- 0.00660✓
- MGE
- 0.02841✓
- MAE
- 0.00747
- Conn.
- 0.00708
- MGE
- 0.05037
- MAE
- 0.01459✓
- Conn.
- 0.01285✓
- MGE
- 0.04911✓
- MAE
- 0.01525
- Conn.
- 0.01362
- MGE
- 0.03855✓
- MAE
- 0.01098✓
- Conn.
- 0.01085✓
- MGE
- 0.03878
- MAE
- 0.01145
- Conn.
- 0.01170
- MGE
- 0.08535
- MAE
- 0.01467
- Conn.
- 0.01374
- MGE
- 0.08066✓
- MAE
- 0.01395✓
- Conn.
- 0.01311✓
- MGE
- 0.02484✓
- MAE
- 0.00507
- Conn.
- 0.00359
- MGE
- 0.02619
- MAE
- 0.00475✓
- Conn.
- 0.00359✓
- MGE
- 0.03527
- MAE
- 0.01145✓
- Conn.
- 0.01039
- MGE
- 0.03370✓
- MAE
- 0.01148
- Conn.
- 0.01015✓
- MGE
- 0.05925✓
- MAE
- 0.01248✓
- Conn.
- 0.01189✓
- MGE
- 0.07350
- MAE
- 0.01472
- Conn.
- 0.01452
- MGE
- 0.04499✓
- MAE
- 0.00897✓
- Conn.
- 0.00783✓
- MGE
- 0.04902
- MAE
- 0.00965
- Conn.
- 0.00845
- MGE
- 0.01596
- MAE
- 0.00448
- Conn.
- 0.00366
- MGE
- 0.01263✓
- MAE
- 0.00370✓
- Conn.
- 0.00266✓
- MGE
- 0.04262✓
- MAE
- 0.00947
- Conn.
- 0.00933✓
- MGE
- 0.04433
- MAE
- 0.00942✓
- Conn.
- 0.00934
- MGE
- 0.08021
- MAE
- 0.01758
- Conn.
- 0.01680
- MGE
- 0.07464✓
- MAE
- 0.01590✓
- Conn.
- 0.01520✓
- MGE
- 0.01995
- MAE
- 0.00915
- Conn.
- 0.00867
- MGE
- 0.01940✓
- MAE
- 0.00869✓
- Conn.
- 0.00767✓
- MGE
- 0.01569✓
- MAE
- 0.00931✓
- Conn.
- 0.00761✓
- MGE
- 0.01867
- MAE
- 0.01305
- Conn.
- 0.01152
- MGE
- 0.05683✓
- MAE
- 0.01115✓
- Conn.
- 0.00906✓
- MGE
- 0.08035
- MAE
- 0.02283
- Conn.
- 0.02099
- MGE
- 0.03150
- MAE
- 0.00535
- Conn.
- 0.00470
- MGE
- 0.02727✓
- MAE
- 0.00459✓
- Conn.
- 0.00401✓
- MGE
- 0.31724✓
- MAE
- 0.28385✓
- Conn.
- 0.28757✓
- MGE
- 0.34795
- MAE
- 0.46062
- Conn.
- 0.47012
- MGE
- 0.02866✓
- MAE
- 0.00615✓
- Conn.
- 0.00474✓
- MGE
- 0.03023
- MAE
- 0.00675
- Conn.
- 0.00548
- MGE
- 0.00923✓
- MAE
- 0.00274✓
- Conn.
- 0.00143✓
- MGE
- 0.01159
- MAE
- 0.00830
- Conn.
- 0.00688
- MGE
- 0.01942
- MAE
- 0.00753✓
- Conn.
- 0.00536
- MGE
- 0.01939✓
- MAE
- 0.00754
- Conn.
- 0.00501✓
- MGE
- 0.05859
- MAE
- 0.00997
- Conn.
- 0.00911
- MGE
- 0.05645✓
- MAE
- 0.00930✓
- Conn.
- 0.00850✓
- MGE
- 0.03481
- MAE
- 0.02009
- Conn.
- 0.01960
- MGE
- 0.03070✓
- MAE
- 0.01848✓
- Conn.
- 0.01790✓
- MGE
- 0.01605
- MAE
- 0.00343
- Conn.
- 0.00229
- MGE
- 0.01599✓
- MAE
- 0.00337✓
- Conn.
- 0.00219✓
- MGE
- 0.01019✓
- MAE
- 0.00325✓
- Conn.
- 0.00190✓
- MGE
- 0.01045
- MAE
- 0.00402
- Conn.
- 0.00262
- MGE
- 0.02151
- MAE
- 0.00860
- Conn.
- 0.00740
- MGE
- 0.02125✓
- MAE
- 0.00843✓
- Conn.
- 0.00666✓
- MGE
- 0.03805✓
- MAE
- 0.01183✓
- Conn.
- 0.01043✓
- MGE
- 0.05709
- MAE
- 0.02184
- Conn.
- 0.02052
- MGE
- 0.05077✓
- MAE
- 0.03474✓
- Conn.
- 0.03567✓
- MGE
- 0.08856
- MAE
- 0.06629
- Conn.
- 0.06887
- MGE
- 0.40325✓
- MAE
- 0.91719
- Conn.
- 0.92302
- MGE
- 0.45101
- MAE
- 0.90677✓
- Conn.
- 0.91173✓
- MGE
- 0.14172
- MAE
- 0.10020
- Conn.
- 0.09928
- MGE
- 0.12912✓
- MAE
- 0.08314✓
- Conn.
- 0.08233✓
- MGE
- 0.13010
- MAE
- 0.02335
- Conn.
- 0.02242
- MGE
- 0.12194✓
- MAE
- 0.02224✓
- Conn.
- 0.02147✓
- MGE
- 0.06450
- MAE
- 0.05163✓
- Conn.
- 0.05022✓
- MGE
- 0.06271✓
- MAE
- 0.21489
- Conn.
- 0.21447
- MGE
- 0.03602
- MAE
- 0.01193
- Conn.
- 0.01118
- MGE
- 0.03058✓
- MAE
- 0.00983✓
- Conn.
- 0.00802✓
- MGE
- 0.05163
- MAE
- 0.01897
- Conn.
- 0.01740
- MGE
- 0.04706✓
- MAE
- 0.01840✓
- Conn.
- 0.01632✓
- MGE
- 0.01632✓
- MAE
- 0.00735
- Conn.
- 0.00545
- MGE
- 0.01689
- MAE
- 0.00690✓
- Conn.
- 0.00444✓
Methodology
- Alpha mattes are grayscale masks encoding per-pixel foreground probability. They capture fine edge detail (hair, fur, semi-transparent regions) that binary masks discard.
- Metrics are computed by comparing predicted mattes against a ground-truth reference: MGE measures edge sharpness, MAE measures pixel-level accuracy, and Connectivity penalises fragmented foreground regions. All three are lower-is-better.
- Scores are shown as ? when no ground-truth matte is available for that image. See the alpha matting evaluation benchmark for details on the evaluation methodology.