Papers › Affordance Transfer Learning for Human-Object Interaction Detection

Affordance Transfer Learning for Human-Object Interaction Detection

7 Apr 2021CVPR 2021 1arXiv:2104.02867archive 2025-07-28

Zhi Hou, Baosheng Yu, Yu Qiao, Xiaojiang Peng, DaCheng Tao

Reasoning the human-object interactions (HOI) is essential for deeper scene understanding, while object affordances (or functionalities) are of great importance for human to discover unseen HOIs with novel objects. Inspired by this, we introduce an affordance transfer learning approach to jointly detect HOIs with novel objects and recognize affordances. Specifically, HOI representations can be decoupled into a combination of affordance and object representations, making it possible to compose novel interactions by combining affordance representations and novel object representations from additional images, i.e. transferring the affordance to novel objects. With the proposed affordance transfer learning, the model is also capable of inferring the affordances of novel objects from known affordance representations. The proposed method can thus be used to 1) improve the performance of HOI detection, especially for the HOIs with unseen objects; and 2) infer the affordances of novel objects. Experimental results on two datasets, HICO-DET and HOI-COCO (from V-COCO), demonstrate significant improvements over recent state-of-the-art methods for HOI detection and object affordance detection. Code is available at https://github.com/zhihou7/HOI-CL

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

zhihou7/HOI-CL officialmentioned in papermentioned on GitHubtf report
zhihou7/HOI-CL-OneStage officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Affordance DetectionAffordance RecognitionHuman-Object Interaction Concept DiscoveryHuman-Object Interaction DetectionObjectScene UnderstandingTransfer Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Affordance Recognition HICO-DET ATL COCO-Val2017 52.01 #2 of 4 Archive leaderboard report
Affordance Recognition HICO-DET ATL HICO 59.44 #2 of 4 Archive leaderboard report
Affordance Recognition HICO-DET ATL Novel classes 15.64 #2 of 4 Archive leaderboard report
Affordance Recognition HICO-DET ATL Object365 50.94 #2 of 4 Archive leaderboard report
Affordance Recognition HICO-DET(Unknown Concepts) ATL COCO-Val2017 36.80 #2 of 3 Archive leaderboard report
Affordance Recognition HICO-DET(Unknown Concepts) ATL HICO 42.00 #2 of 3 Archive leaderboard report
Affordance Recognition HICO-DET(Unknown Concepts) ATL Novel Classes 15.64 #2 of 3 Archive leaderboard report
Affordance Recognition HICO-DET(Unknown Concepts) ATL Obj365 34.38 #2 of 3 Archive leaderboard report
Human-Object Interaction Concept Discovery HICO-DET Affordance Transfer Unknown (AP) 24.38 #3 of 3 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections