AI Paper Atlas

Research map / Video / Video-language understanding

COMPOSER: Compositional Reasoning of Group Activity in Videos with Keypoint-Only Modality

Honglu Zhou, Asim Kadav, Aviv Shamsian, Shijie Geng, Farley Lai, Long Zhao, Ting Liu, Mubbasir Kapadia, Hans Peter Graf

ECCV 2022 · Video › Video-language understanding · approach: Event and anomaly reasoning in video · 42 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
use keypoint-only input instead of RGB features (ablation)COMPOSER RGB variant · Volleyball Olympic splitgroup activity accuracy (%): 81.1 -> 95.1Improved
add contrastive clustering for scale agreement (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 93.4 -> 94.6Improved
use contrastive clustering instead of label consistency (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 93.9 -> 94.6Improved
increase number of scales from 1 to 4 (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 91.2 -> 94.6Improved
add auxiliary predictions at each scale and block (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 93.3 -> 94.6Improved
add Multiscale Transformer instead of a flat model (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 88.1 -> 94.6Improved
add Actor Dropout augmentation (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 94.0 -> 94.6Improved
add horizontal flip augmentation (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 94.0 -> 94.6Improved
add horizontal and vertical move augmentation (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 94.2 -> 94.6Improved
fine-to-coarse token order instead of reverse (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 86.8 -> 94.6Improved
separate transformer encoder per scale instead of one encoder for all tokens (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 92.4 -> 94.6Improved
use fixed person grouping instead of time-varying grouping (ablation)COMPOSER · Volleyball Original splitgroup activity accuracy (%): 92.8 -> 94.6Improved

Later papers that compared with this method

1 later accepted paper reports COMPOSER as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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