SOTAVerified

Human-Object Interaction Detection

Human-Object Interaction (HOI) detection is a task of identifying "a set of interactions" in an image, which involves the i) localization of the subject (i.e., humans) and target (i.e., objects) of interaction, and ii) the classification of the interaction labels.

Papers

Showing 5175 of 449 papers

TitleStatusHype
HOI Analysis: Integrating and Decomposing Human-Object InteractionCode1
HOICLIP: Efficient Knowledge Transfer for HOI Detection with Vision-Language ModelsCode1
BEHAVE: Dataset and Method for Tracking Human Object InteractionsCode1
DRG: Dual Relation Graph for Human-Object Interaction DetectionCode1
Efficient Two-Stage Detection of Human-Object Interactions with a Novel Unary-Pairwise TransformerCode1
IMos: Intent-Driven Full-Body Motion Synthesis for Human-Object InteractionsCode1
ConsNet: Learning Consistency Graph for Zero-Shot Human-Object Interaction DetectionCode1
Ins-HOI: Instance Aware Human-Object Interactions RecoveryCode1
End-to-End Human Object Interaction Detection with HOI TransformerCode1
Interaction Compass: Multi-Label Zero-Shot Learning of Human-Object Interactions via Spatial RelationsCode1
Discovering Human-Object Interaction Concepts via Self-Compositional LearningCode1
DIRV: Dense Interaction Region Voting for End-to-End Human-Object Interaction DetectionCode1
Disentangled Pre-training for Human-Object Interaction DetectionCode1
Detailed 2D-3D Joint Representation for Human-Object InteractionCode1
Boosting Human-Object Interaction Detection with Text-to-Image Diffusion ModelCode1
Detecting Human-Object Interactions with Action Co-occurrence PriorsCode1
Diagnosing Human-object Interaction DetectorsCode1
D3D-HOI: Dynamic 3D Human-Object Interactions from VideosCode1
Discovering A Variety of Objects in Spatio-Temporal Human-Object InteractionsCode1
Discovering Human Interactions With Novel Objects via Zero-Shot LearningCode1
DECO: Dense Estimation of 3D Human-Scene Contact In The WildCode1
Affordance Transfer Learning for Human-Object Interaction DetectionCode1
Cascaded Human-Object Interaction RecognitionCode1
Dynamics-Regulated Kinematic Policy for Egocentric Pose EstimationCode1
Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object InteractionsCode1
Show:102550
← PrevPage 3 of 18Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Ours (PViC+)mAP46.49Unverified
2RLIPv2 (Swin-L)mAP45.09Unverified
3PViC-SwinLmAP44.32Unverified
4SOV-STG (Swin-L)mAP43.35Unverified
5DiffHOImAP41.5Unverified
6ViPLOmAP37.22Unverified
7FGAHOImAP37.18Unverified
8ERNetmAP36.89Unverified
9CQL+GEN-VLKT-LmAP36.03Unverified
10QAHOI (Swin-L)mAP35.78Unverified
#ModelMetricClaimedVerifiedStatus
1RLIPv2AP(S1)72.1Unverified
2MURENAP(S1)68.8Unverified
3STIPAP(S1)66Unverified
4DiffHOIAP(S1)65.7Unverified
5OCN (ResNet101)AP(S1)65.3Unverified
6OCN (ResNet50)AP(S1)64.2Unverified
7CDN (ResNet101)AP(S1)63.91Unverified
8HOICLIPAP(S1)63.5Unverified
9QPIC + CPCMAP63.1Unverified
10Body Part InteractivenessAP(S1)63Unverified
#ModelMetricClaimedVerifiedStatus
1DEFRmAP65.6Unverified
2HAKEmAP47.1Unverified
3PaStaNetmAP46.3Unverified
4RelViTmAP43.98Unverified
5Pairwise-PartmAP39.9Unverified
6Mallya & LazebnikmAP36.1Unverified
7Girdhar & RamananmAP34.6Unverified
8R*CNNmAP28.5Unverified
#ModelMetricClaimedVerifiedStatus
1HOI4ABOTDetection: Full (mAP@0.5)11.12Unverified
2ST-GAZEDetection: Full (mAP@0.5)10.4Unverified
3STTRANDetection: Full (mAP@0.5)7.61Unverified
#ModelMetricClaimedVerifiedStatus
1DJ-RNmAP10.37Unverified
2iCANmAP8.14Unverified
#ModelMetricClaimedVerifiedStatus
1SlowFast + FasterRCNNmAP@0.5 role25.93Unverified