List your company
ResearchFormerly WizWand

Human-Object Interaction Detection on V-COCO (AP Role Scenarios)

73.6AP Role (Scenario 1)InCoM-Net
44.16851.80959.4567.091Sep 16, 2025Oct 24, 2025Dec 2, 2025Jan 10, 2026Feb 17, 2026Mar 28, 2026May 6, 2026
Updated 5mo ago

Evaluation results

MethodLinks
InCoM-NetBackbone=R50+CLIP, Vision Encoder=ViT-L2026.0473.675.4
InCoM-NetBackbone=R50+CLIP, Vision Encoder=ViT-B2026.0472.274.2
MGNMLBackbone=Swin-L+CLIP, Architecture Style=Two-stage2025.0971.274.5
NMSRBackbone=R50+CLIP, Vision Encoder=ViT-L2026.0469.872.1
HORP-LBackbone=Swin-L+CLIP+Gaze, Architecture Style=Two-stage2025.0968.971.1
MGNMBackbone=R50+CLIP, Architecture Style=Two-stage2025.0968.970.6
RLIPv2Backbone=Swin-T+BLIP, Architecture Style=Two-stage2025.0968.870.8
SPLHOIBackbone=R50+BLIP2, Vision Encoder=ViT-L2026.0468.172.9
ScriptHOI2026.0567.869.4
InterProDaBackbone=R50+CLIP, Vision Encoder=ViT-L2026.0467.6—
SCTCBackbone=R50+CLIP, Vision Encoder=ViT-B2026.0467.171.7
GroupHOIBackbone=R50+CLIP, Vision Encoder=ViT-L2026.0466.467.3
MP-HOIBackbone=R50+CLIP+SD, Architecture Style=One-stage2025.0966.267.6
ContextHOIBackbone=R50+CLIP, Vision Encoder=ViT-L2026.0466.1—
STIPBackbone=R502026.046670.7
CLIP4HOI2026.056667.5
BCOMBackbone=R50+CLIP, Vision Encoder=ViT-L2026.0465.869.9
DiffHOI-LBackbone=Swin-L+CLIP+SD, Architecture Style=One-stage2025.0965.768.2
HORPBackbone=R50+CLIP+Gaze, Architecture Style=Two-stage2025.0965.668.3
HORPBackbone=R50+CLIP, Vision Encoder=ViT-B2026.0465.668.3
UniHOIBackbone=R50+BLIP2, Vision Encoder=ViT-L2026.0465.668.3
GEN-VLKT2026.0565.667.2
GroupHOIBackbone=R50+CLIP, Vision Encoder=ViT-B2026.046566
PViC-LBackbone=Swin-L, Architecture Style=Two-stage2025.0964.170.2
CDN2026.0563.965.9
RmLRBackbone=R502026.0463.869.8
DOQBackbone=R50+CLIP, Architecture Style=One-stage2025.0963.5—
HOICLIPBackbone=R50+CLIP, Vision Encoder=ViT-B2026.0463.564.8
ILCNBackbone=Swin-T, Architecture Style=Two-stage2025.0963.465.3
LAINBackbone=R50+CLIP, Architecture Style=Two-stage2025.0963.465.3
Pose-awareBackbone=Swin-L+ViTPose, Architecture Style=Two-stage2025.096368.7
PViCBackbone=R50, Architecture Style=Two-stage2025.0962.867.8
PViCBackbone=R502026.0462.867.8
GEN-VLKTBackbone=R50+CLIP, Vision Encoder=ViT-B2026.0462.464.5
CDTBackbone=R50, Architecture Style=One-stage2025.0961.465.4
DiffHOIBackbone=R50+CLIP+SD, Architecture Style=One-stage2025.0961.163.5
ViPLOBackbone=R101-DC+CLIP+ViTPose, Architecture Style=Two-stage2025.0960.966.6
UPTBackbone=R101, Architecture Style=Two-stage2025.0960.766.2
VDRPBackbone=R50+CLIP, Vision Encoder=ViT-L2026.0460.666.2
FGAHOIBackbone=Swin-T, Architecture Style=One-stage2025.0960.561.2
HOLaBackbone=R50+CLIP, Vision Encoder=ViT-L2026.0460.366
UPTBackbone=R502026.045964.5
QPICBackbone=R502026.0458.861
QPIC2026.0558.861
ADA-CMBackbone=R50+CLIP, Vision Encoder=ViT-L2026.0458.664
ADA-CMBackbone=R50+CLIP, Architecture Style=Two-stage2025.0956.161.5
HOTR2026.0555.264.4
SCGBackbone=R50, Architecture Style=Two-stage2025.0954.260.9
AS-NetBackbone=R50, Architecture Style=One-stage2025.0953.9—
DRG2026.055156.4
VCL2026.0548.354.7
UnionDetBackbone=R50, Architecture Style=One-stage2025.0947.556.2
iCAN2026.0545.352.4
CLIP4HOIBackbone=R50+CLIP, Vision Encoder=ViT-B2026.04—66.3
CMMPBackbone=R50+CLIP, Vision Encoder=ViT-L2026.04—64
LAINBackbone=R50+CLIP, Vision Encoder=ViT-B2026.04—65.1