AI Paper Atlas

Research map / Video / Video segmentation and tracking

HAMoBE: Hierarchical and Adaptive Mixture of Biometric Experts for Video-based Person ReID

Yiyang Su, Yunping Shi, Feng Liu, Xiaoming Liu

ICCV 2025 · Video › Video segmentation and tracking · approach: Spatial-temporal video person re-identification · 6 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
replace vanilla MoE with hierarchical biometric experts (no consistency losses) (ablation)HAMoBE · MEVID same clothestop-1 (%): 63.0 -> 68.1Improved
replace vanilla MoE with hierarchical biometric experts (no consistency losses) (ablation)HAMoBE · MEVID different clothestop-1 (%): 14.3 -> 13.5Hurt
add consistency losses and contrastive loss (ablation)HAMoBE · MEVID same clothestop-1 (%): 68.1 -> 74.0Improved
add consistency losses and contrastive loss (ablation)HAMoBE · MEVID different clothestop-1 (%): 13.5 -> 20.1Improved
add dual-input decision mechanism to single-input mode (ablation)HAMoBE · CCVID different clothestop-1 (%): 85.7 -> 87.6Improved
combine long-term, short-term and temporal experts instead of temporal expert alone (ablation)HAMoBE · CCVID different clothestop-1 (%): 85.4 -> 87.6Improved
increase first-layer experts from 4 to 8 (ablation)HAMoBE · CCVID different clothestop-1 (%): 82.6 -> 85.8Improved
increase first-layer experts from 8 to 16 (ablation)HAMoBE · CCVID different clothestop-1 (%): 85.8 -> 84.5Hurt
use Transformer decoder instead of Transformer encoder for temporal aggregation (ablation)HAMoBE temporal expert · MEVID different clothestop-1 (%): 15.9 -> 20.1Improved
use Transformer decoder instead of mean pooling for temporal aggregation (ablation)HAMoBE temporal expert · MEVID different clothestop-1 (%): 9.8 -> 20.1Improved
use HAMoBE instead of TF-CLIPHAMoBE · LS-VIDmAP (%): 83.8 -> 85.2Improved

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