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

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 411–420 of 449 papers

Title	Date	Tasks	Status	Hype
Detecting Human-Object Interactions via Functional Generalization	Apr 5, 2019	Human-Object Interaction DetectionObject	—Unverified	0
Do Deep Neural Networks Model Nonlinear Compositionality in the Neural Representation of Human-Object Interactions?	Mar 31, 2019	Human-Object Interaction DetectionObject	—Unverified	0
Turbo Learning Framework for Human-Object Interactions Recognition and Human Pose Estimation	Mar 15, 2019	Human-Object Interaction DetectionPose Estimation	—Unverified	0
Grounded Human-Object Interaction Hotspots from Video	Dec 11, 2018	Human-Object Interaction DetectionObject	CodeCode Available	0
Transferable Interactiveness Knowledge for Human-Object Interaction Detection	Nov 20, 2018	Human-Object Interaction DetectionObject	CodeCode Available	1
No-Frills Human-Object Interaction Detection: Factorization, Layout Encodings, and Training Techniques	Nov 14, 2018	Human-Object Interaction DetectionObject	CodeCode Available	0
Compositional Learning for Human Object Interaction	Sep 1, 2018	Human-Object Interaction DetectionObject	—Unverified	0
iCAN: Instance-Centric Attention Network for Human-Object Interaction Detection	Aug 30, 2018	Human-Object Interaction DetectionObject	CodeCode Available	1
Interact as You Intend: Intention-Driven Human-Object Interaction Detection	Aug 29, 2018	Human-Object Interaction Detection	—Unverified	0
Learning Human-Object Interactions by Graph Parsing Neural Networks	Aug 23, 2018	Human-Object Interaction DetectionObject	CodeCode Available	0

Show:10 25 50

← PrevPage 42 of 45Next →

All datasets HICO-DET V-COCO HICO VidHOI Ambiguious-HOI MECCANO

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	Ours (PViC+)	mAP	46.49	—	Unverified
2	RLIPv2 (Swin-L)	mAP	45.09	—	Unverified
3	PViC-SwinL	mAP	44.32	—	Unverified
4	SOV-STG (Swin-L)	mAP	43.35	—	Unverified
5	DiffHOI	mAP	41.5	—	Unverified
6	ViPLO	mAP	37.22	—	Unverified
7	FGAHOI	mAP	37.18	—	Unverified
8	ERNet	mAP	36.89	—	Unverified
9	CQL+GEN-VLKT-L	mAP	36.03	—	Unverified
10	QAHOI (Swin-L)	mAP	35.78	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	RLIPv2	AP(S1)	72.1	—	Unverified
2	MUREN	AP(S1)	68.8	—	Unverified
3	STIP	AP(S1)	66	—	Unverified
4	DiffHOI	AP(S1)	65.7	—	Unverified
5	OCN (ResNet101)	AP(S1)	65.3	—	Unverified
6	OCN (ResNet50)	AP(S1)	64.2	—	Unverified
7	CDN (ResNet101)	AP(S1)	63.91	—	Unverified
8	HOICLIP	AP(S1)	63.5	—	Unverified
9	QPIC + CPC	MAP	63.1	—	Unverified
10	Body Part Interactiveness	AP(S1)	63	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DEFR	mAP	65.6	—	Unverified
2	HAKE	mAP	47.1	—	Unverified
3	PaStaNet	mAP	46.3	—	Unverified
4	RelViT	mAP	43.98	—	Unverified
5	Pairwise-Part	mAP	39.9	—	Unverified
6	Mallya & Lazebnik	mAP	36.1	—	Unverified
7	Girdhar & Ramanan	mAP	34.6	—	Unverified
8	R*CNN	mAP	28.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	HOI4ABOT	Detection: Full (mAP@0.5)	11.12	—	Unverified
2	ST-GAZE	Detection: Full (mAP@0.5)	10.4	—	Unverified
3	STTRAN	Detection: Full (mAP@0.5)	7.61	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DJ-RN	mAP	10.37	—	Unverified
2	iCAN	mAP	8.14	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SlowFast + FasterRCNN	mAP@0.5 role	25.93	—	Unverified