Upload accuracy.csv with huggingface_hub
Browse files- accuracy.csv +7 -2
accuracy.csv
CHANGED
|
@@ -79,12 +79,17 @@ pyfeat_v2,eyediap,gaze,angular_mae_deg,19.975,800,out_of_sample,"a=pitch,b=yaw;s
|
|
| 79 |
pyfeat_v2,eyediap_norm,gaze,angular_mae_deg,19.418,800,out_of_sample,"a=pitch,b=yaw;sign_pitch=-1;sign_yaw=1;unit=rad"
|
| 80 |
pyfeat_v2,gaze360,gaze,angular_mae_deg,9.416,2425,held_out,"a=pitch,b=yaw;sign_pitch=1;sign_yaw=1;unit=rad"
|
| 81 |
pyfeat_v2,mpiifacegaze,gaze,angular_mae_deg,2.802,3000,in_sample,"a=pitch,b=yaw;sign_pitch=1;sign_yaw=-1;unit=rad"
|
| 82 |
-
pyfeat_v2,mpiigaze,gaze,reported_mae_deg,
|
| 83 |
-
pyfeat_v2,gaze360,gaze,reported_mae_deg,
|
| 84 |
openface3,mpiifacegaze,gaze,reported_mae_deg,2.56,,in_sample,OpenFace 3.0 paper (arXiv:2506.02891)
|
| 85 |
openface3,gaze360,gaze,reported_mae_deg,10.6,,in_sample,OpenFace 3.0 paper (arXiv:2506.02891)
|
| 86 |
l2cs_net,mpiigaze,gaze,reported_mae_deg,3.92,,unknown,L2CS-Net paper (arXiv:2203.03339); py-feat's gaze lineage
|
| 87 |
l2cs_net,gaze360,gaze,reported_mae_deg,10.41,,unknown,L2CS-Net paper (arXiv:2203.03339); py-feat's gaze lineage
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
openface3,affectnet_val,emotion,accuracy,0.4929,980,held_out,
|
| 89 |
openface3,affectnet_val,emotion,f1_macro,0.5198,980,held_out,
|
| 90 |
openface3,raf_test,emotion,accuracy,0.5132,2233,out_of_sample,
|
|
|
|
| 79 |
pyfeat_v2,eyediap_norm,gaze,angular_mae_deg,19.418,800,out_of_sample,"a=pitch,b=yaw;sign_pitch=-1;sign_yaw=1;unit=rad"
|
| 80 |
pyfeat_v2,gaze360,gaze,angular_mae_deg,9.416,2425,held_out,"a=pitch,b=yaw;sign_pitch=1;sign_yaw=1;unit=rad"
|
| 81 |
pyfeat_v2,mpiifacegaze,gaze,angular_mae_deg,2.802,3000,in_sample,"a=pitch,b=yaw;sign_pitch=1;sign_yaw=-1;unit=rad"
|
| 82 |
+
pyfeat_v2,mpiigaze,gaze,reported_mae_deg,4.24,,unknown,"au_deep v2.4 (shipped ckpt, v24_s3 run; self-reported, in-distribution)"
|
| 83 |
+
pyfeat_v2,gaze360,gaze,reported_mae_deg,9.4,,held_out,"au_deep v2.4 (shipped ckpt, v24_s3 run; self-reported, in-distribution; deployed ckpt stores gaze_mae 9.9)"
|
| 84 |
openface3,mpiifacegaze,gaze,reported_mae_deg,2.56,,in_sample,OpenFace 3.0 paper (arXiv:2506.02891)
|
| 85 |
openface3,gaze360,gaze,reported_mae_deg,10.6,,in_sample,OpenFace 3.0 paper (arXiv:2506.02891)
|
| 86 |
l2cs_net,mpiigaze,gaze,reported_mae_deg,3.92,,unknown,L2CS-Net paper (arXiv:2203.03339); py-feat's gaze lineage
|
| 87 |
l2cs_net,gaze360,gaze,reported_mae_deg,10.41,,unknown,L2CS-Net paper (arXiv:2203.03339); py-feat's gaze lineage
|
| 88 |
+
libreface,gaze360,gaze,reported_mae_deg,14.68,,out_of_sample,"LibreFace 2.0 landmark-MLP, 'Is Geometry Enough?' (arXiv:2603.24724); within-domain; landmark-based"
|
| 89 |
+
lit_sota_within,eyediap,gaze,reported_mae_deg,5.0,,unknown,"EYEDIAP within-dataset SOTA: MAFI-Gaze 5.0, RT-Gene 6.3 (general gaze nets, not our tools)"
|
| 90 |
+
lit_sota_crossds,eyediap,gaze,reported_mae_deg,8.1,,unknown,"EYEDIAP cross-dataset zero-shot: GMMGaze 8.1, GazeNet 9.6 (general gaze nets)"
|
| 91 |
+
lit_sota_within,mpiigaze,gaze,reported_mae_deg,3.3,,unknown,MPIIGaze within-dataset SOTA range ~3-4 deg (appearance-based)
|
| 92 |
+
lit_sota_within,gaze360,gaze,reported_mae_deg,10.0,,unknown,Gaze360 within-dataset SOTA ~10 deg (original baseline); ~9 with recent nets
|
| 93 |
openface3,affectnet_val,emotion,accuracy,0.4929,980,held_out,
|
| 94 |
openface3,affectnet_val,emotion,f1_macro,0.5198,980,held_out,
|
| 95 |
openface3,raf_test,emotion,accuracy,0.5132,2233,out_of_sample,
|