AI Paper Atlas

Research map / Video / Video segmentation and tracking

AFF-ttention! Affordances and Attention models for Short-Term Object Interaction Anticipation

Lorenzo Mur-Labadia, Ruben Martinez-Cantin, Jose J Guerrero, Giovanni Maria Farinella, Antonino Furnari

ECCV 2024 · Video › Video segmentation and tracking · approach: Physics-inspired motion prediction · 13 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 environment affordance (E.AFF) fusion (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP N+V: 10.75 -> 11.75Improved
add interaction hotspot confidence re-weighting (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP N: 21.71 -> 23.63Improved
add both affordances and hotspots (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP All: 3.53 -> 3.77Improved
add both affordance modules (ablation)StillFast · Ego4D-STA v1 valTop-5 mAP All: 2.48 -> 2.95Improved
use EGO-TOPO trained NN classifier for affordance distribution instead of kNN zone matching (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP All: 3.74 -> 2.64Hurt
replace mean temporal pooling with frame-guided temporal pooling attention (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP All: 2.63 -> 3.39Improved
replace sum fusion with dual image-video attention (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP All: 3.39 -> 3.47Improved
use one-directional cross-attention instead of dual cross-attention (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP All: 3.53 -> 3.01 (I to V); 3.35 (V to I)Hurt
extract per-frame DINOv2 video features with mean pooling instead of a video model (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP All: 2.52 -> 2.19Hurt
fine-tune last blocks of DINOv2 image encoder (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP N: 20.08 -> 21.71Improved
use multi-head attention in temporal pooling and fusion (ablation)STAformer · Ego4D-STA v1 valTop-5 mAP All: 3.53 -> 3.85Improved
train on Ego4D v2 instead of v1STAformer · Ego4D-STA testTop-5 mAP All: 4.06 -> 6.75Improved

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