AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Spatially Conditioned Graphs for Detecting Human-Object Interactions

Frederic Z. Zhang, Dylan Campbell, Stephen Gould

ICCV 2021 · Detection and segmentation › Open-set and few-shot detection · approach: Visual relationship and scene graph detection · 143 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add spatial conditioning at all stages (adjacency, messages, global features, refinement) (ablation)appearance-only bipartite graph · HICO-DET, fine-tuned detectormAP: 28.43 -> 31.33Improved
add spatial conditioning at all stages (ablation)appearance-only bipartite graph · HICO-DET, COCO detectormAP: 19.92 -> 21.85Improved
add spatial conditioning to the messages only (ablation)appearance-only graph · HICO-DET, fine-tuned detectormAP: 28.43 -> 30.06Improved
add spatial conditioning to the adjacency only (ablation)appearance-only graph · HICO-DET, fine-tuned detectormAP: 28.43 -> 29.48Improved
add spatial conditioning to the refinement only (ablation)appearance-only graph · HICO-DET, COCO detectormAP: 19.92 -> 21.03Improved
increase multi-branch fusion cardinality from 1 to 16 (product) (ablation)MBF module · HICO-DET, fine-tuned detectormAP: 30.75 -> 31.33Improved
increase cardinality from 1 to 16 (product) (ablation)MBF module · HICO-DET, COCO detectormAP: 21.18 -> 21.85Improved
replace elementwise product with sum in fusion (c=16) (ablation)MBF module · HICO-DET, fine-tuned detectormAP: 31.33 -> 31.07No clear effect
replace elementwise product with concatenation (c=16) (ablation)MBF module · HICO-DET, fine-tuned detectormAP: 31.33 -> 31.65No clear effect
add message passing (T=0 to T=2) (ablation)spatially conditioned graph · HICO-DET, fine-tuned detectormAP: 28.86 -> 31.33Improved
add a third message passing iteration (T=2 to T=3) (ablation)spatially conditioned graph · HICO-DETmAP: 21.85 -> 21.72 (COCO); 31.33 -> 31.78 (fine-tuned)Mixed
improve detection quality (COCO to fine-tuned to oracle) for gain of spatial branch over appearance-onlyfull model vs appearance-only baseline · HICO-DETmAP difference: +1.93 -> +2.90 -> +4.36Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy