Human Object Interaction Detection On V Coco

评估指标

AP(S1)

AP(S2)

Time Per Frame(ms)

评测结果

各个模型在此基准测试上的表现结果

				Paper Title
RLIPv2	72.1	74.1	-	RLIPv2: Fast Scaling of Relational Language-Image Pre-training
MUREN	68.8	71.0	-	Relational Context Learning for Human-Object Interaction Detection
STIP	66.0	70.7	74	Exploring Structure-aware Transformer over Interaction Proposals for Human-Object Interaction Detection
DiffHOI	65.7	68.2	-	Boosting Human-Object Interaction Detection with Text-to-Image Diffusion Model
OCN (ResNet101)	65.3	67.1	-	Detecting Human-Object Interactions with Object-Guided Cross-Modal Calibrated Semantics
OCN (ResNet50)	64.2	66.3	43	Detecting Human-Object Interactions with Object-Guided Cross-Modal Calibrated Semantics
CDN (ResNet101)	63.91	65.89	-	Mining the Benefits of Two-stage and One-stage HOI Detection
HOICLIP	63.50	64.81	-	HOICLIP: Efficient Knowledge Transfer for HOI Detection with Vision-Language Models
Body Part Interactiveness	63.0	65.1	-	Mining Cross-Person Cues for Body-Part Interactiveness Learning in HOI Detection
UPT-R101-DC5	61.3	67.1	131	Efficient Two-Stage Detection of Human-Object Interactions with a Novel Unary-Pairwise Transformer
UPT-R101	60.7	66.2	64	Efficient Two-Stage Detection of Human-Object Interactions with a Novel Unary-Pairwise Transformer
UPT-R50	59.0	64.5	43	Efficient Two-Stage Detection of Human-Object Interactions with a Novel Unary-Pairwise Transformer
QPIC (ResNet50)	58.8	61.0	46	QPIC: Query-Based Pairwise Human-Object Interaction Detection with Image-Wide Contextual Information
QPIC (ResNet101)	58.3	60.7	63	QPIC: Query-Based Pairwise Human-Object Interaction Detection with Image-Wide Contextual Information
DIRV	56.1	-	68	DIRV: Dense Interaction Region Voting for End-to-End Human-Object Interaction Detection
HOTR	55.2	64.4	-	HOTR: End-to-End Human-Object Interaction Detection with Transformers
GGNet	54.7	-	-	Glance and Gaze: Inferring Action-aware Points for One-Stage Human-Object Interaction Detection
SCG	54.2	60.9	500	Spatially Conditioned Graphs for Detecting Human-Object Interactions
PD-Net	53.34	-	-	Polysemy Deciphering Network for Robust Human-Object Interaction Detection
IDN	53.3	60.3	-	HOI Analysis: Integrating and Decomposing Human-Object Interaction

0 of 34 row(s) selected.