List your company
ResearchFormerly WizWand

Video Part Segmentation on VIP

0.408mIoUiBOT +Ours
0.249920.290960.3320.37304Mar 19, 2025May 20, 2025Jul 21, 2025Sep 21, 2025Nov 22, 2025Jan 23, 2026Mar 27, 2026
Updated 4mo ago

Evaluation results

MethodLinks
iBOT +OursType=Self Distillation, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=+5, Evaluation Protocol=Zero-shot semi-supervised2026.030.408
DINO v2 +OursType=Self Distillation, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=+5, Evaluation Protocol=Zero-shot semi-supervised2026.030.399
MoCo v3 +OursType=Contrastive Learning, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=+5, Evaluation Protocol=Zero-shot semi-supervised2026.030.398
DINO +OursType=Self Distillation, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=+5, Evaluation Protocol=Zero-shot semi-supervised2026.030.398
BLIP +OursType=Contrastive Learning, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=+5, Evaluation Protocol=Zero-shot semi-supervised2026.030.396
iBOTType=Self Distillation, Backbone=ViT-B/16, Pre-training Dataset=INet, Epoch=400, Evaluation Protocol=Zero-shot semi-supervised2026.030.396
CLIP +OursType=Contrastive Learning, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=+5, Evaluation Protocol=Zero-shot semi-supervised2026.030.392
DINOType=Self Distillation, Backbone=ViT-B/16, Pre-training Dataset=INet, Epoch=300, Evaluation Protocol=Zero-shot semi-supervised2026.030.391
T-CoReBackbone=ViT-B/16, Epoch=200, Pre-trained=Kinetics-4002025.030.389
MoCo v3Type=Contrastive Learning, Backbone=ViT-B/16, Pre-training Dataset=INet, Epoch=300, Evaluation Protocol=Zero-shot semi-supervised2026.030.388
DINO v2Type=Self Distillation, Backbone=ViT-B/16, Pre-training Dataset=INet, Epoch=100, Evaluation Protocol=Zero-shot semi-supervised2026.030.384
DINO v2Backbone=ViT-B/16, Epoch=200, Pre-trained=Kinetics-4002025.030.382
CLIPType=Contrastive Learning, Backbone=ViT-B/16, Pre-training Dataset=WIT, Epoch=32, Evaluation Protocol=Zero-shot semi-supervised2026.030.381
BLIPType=Contrastive Learning, Backbone=ViT-B/16, Pre-training Dataset=LAION, Epoch=20, Evaluation Protocol=Zero-shot semi-supervised2026.030.376
SiamMAEType=Video Pretrained, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=400, Evaluation Protocol=Zero-shot semi-supervised2026.030.361
I-JEPA +OursType=Mask Modeling, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=+5, Evaluation Protocol=Zero-shot semi-supervised2026.030.353
RSPBackbone=ViT-B/16, Epoch=400, Pre-trained=Kinetics-4002025.030.34
ToBoPre-trained=Kinetics-400, Epochs=4002025.070.34
RSPType=Video Pretrained, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=400, Evaluation Protocol=Zero-shot semi-supervised2026.030.34
RSPPre-trained=Kinetics-400, Epochs=4002025.070.338
MAE +OursType=Mask Modeling, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=+5, Evaluation Protocol=Zero-shot semi-supervised2026.030.338
CropMAEPre-trained=Kinetics-400, Epochs=4002025.070.337
DINOPre-trained=Kinetics-400, Epochs=4002025.070.334
DinoType=Non-contrastive, Architecture=ViT-S/16, Pre-training dataset=Kinetics-400, Pre-training epochs=4002025.050.334
SiamMAEPre-trained=Kinetics-400, Epochs=4002025.070.333
SiamMAEType=Autoencoder, Architecture=ViT-S/16, Pre-training dataset=Kinetics-400, Pre-training epochs=4002025.050.333
MAE-STBackbone=ViT-L/16, Epoch=1600, Pre-trained=Kinetics-4002025.030.332
MAE-STType=Video Pretrained, Backbone=ViT-L/16, Pre-training Dataset=K400, Epoch=1600, Evaluation Protocol=Zero-shot semi-supervised2026.030.332
PhiNet v2Backbone=ViT-S/16, Pre-training=Kinetics-400, Pre-training Epochs=4002025.050.331
PhiNet v2Type=Non-contrastive, Architecture=ViT-S/16, Pre-training dataset=Kinetics-400, Pre-training epochs=4002025.050.331
CropMAEBackbone=ViT-B/16, Epoch=400, Pre-trained=Kinetics-4002025.030.33
CropMAEType=Video Pretrained, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=400, Evaluation Protocol=Zero-shot semi-supervised2026.030.33
CropMAEBackbone=ViT-S/16, Pre-training=Kinetics-400, Pre-training Epochs=4002025.050.33
CropMAEType=Autoencoder, Architecture=ViT-S/16, Pre-training dataset=Kinetics-400, Pre-training epochs=4002025.050.33
MAEPre-trained=Kinetics-400, Epochs=4002025.070.325
MAEType=Autoencoder, Architecture=ViT-S/16, Pre-training dataset=Kinetics-400, Pre-training epochs=4002025.050.325
MoCo v3Pre-trained=Kinetics-400, Epochs=4002025.070.324
RSPBackbone=ViT-S/16, Pre-training=Kinetics-400, Pre-training Epochs=4002025.050.324
RSPType=Autoencoder, Architecture=ViT-S/16, Pre-training dataset=Kinetics-400, Pre-training epochs=4002025.050.324
MoCo v3Type=Contrastive, Architecture=ViT-S/16, Pre-training dataset=Kinetics-400, Pre-training epochs=4002025.050.324
SimCLRPre-trained=Kinetics-400, Epochs=4002025.070.319
SimCLRType=Contrastive, Architecture=ViT-S/16, Pre-training dataset=Kinetics-400, Pre-training epochs=4002025.050.319
I-JEPAType=Mask Modeling, Backbone=ViT-B/16, Pre-training Dataset=INet, Epoch=800, Evaluation Protocol=Zero-shot semi-supervised2026.030.315
DropMAEBackbone=ViT-B/16, Epoch=1600, Pre-trained=Kinetics-4002025.030.311
DropMAEType=Video Pretrained, Backbone=ViT-B/16, Pre-training Dataset=K400, Epoch=1600, Evaluation Protocol=Zero-shot semi-supervised2026.030.311
JEPA-like methodBackbone=ViT-S/16, Pre-training=Kinetics-400, Pre-training Epochs=4002025.050.301
MAEType=Mask Modeling, Backbone=ViT-B/16, Pre-training Dataset=INet, Epoch=800, Evaluation Protocol=Zero-shot semi-supervised2026.030.293
VideoMAEType=Video Pretrained, Backbone=ViT-L/16, Pre-training Dataset=K400, Epoch=1600, Evaluation Protocol=Zero-shot semi-supervised2026.030.256