AI Paper Atlas

Research map / Video / Video segmentation and tracking

Orientation Invariant Feature Embedding and Spatial Temporal Regularization for Vehicle Re-Identification

Zhongdao Wang, Luming Tang, Xihui Liu, Zhuliang Yao, Shuai Yi, Jing Shao, Junjie Yan, Shengjin Wang, Hongsheng Li, Xiaogang Wang

ICCV 2017 · Video › Video segmentation and tracking · approach: Spatial-temporal video person re-identification · 373 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 orientation-based region features and aggregation to the single-branch modelBaseline · VeRi-776mAP: 45.50 -> 48.00Improved
add orientation-based region features and aggregation to the single-branch modelBaseline · VehicleID (Large)CMC@1: 63.2 -> 67.0Improved
add spatio-temporal regularization with log-normal transition modelOurs · VeRi-776mAP: 48.00 -> 51.42Improved
add spatio-temporal regularizationOurs · VeRi-776CMC@1: 65.9 -> 68.3Improved
train key point regressor with cross-entropy instead of L2 loss (ablation)key point regressor · VeRi-776 key points, r0=3key point accuracy (%): 87.4 -> 88.8Improved
train key point regressor with cross-entropy instead of L2 loss (ablation)key point regressor · VeRi-776 key points, r0=5key point accuracy (%): 90.50 -> 92.05Improved
use orientation-based region proposals instead of 4 attention masks (ablation)Ours · VeRi-776CMC@1: 64.8 -> 65.9Improved
use orientation-based region proposals instead of 4 attention masks (ablation)Ours · VehicleIDCMC@1: 65.6 -> 66.6Improved
vary number of attention masks from 4 to 8 (ablation)attention network · VeRi-776CMC@1: 64.8 -> 62.7Hurt
add KISSME metric learning on global featuresBaseline · VeRi-776CMC@1: 88.66 -> 89.02Improved
add MLAPG metric learning on global featuresBaseline · VeRi-776CMC@1: 88.66 -> 87.89Hurt
add Zhang et al. metric learning on final featuresOurs · VeRi-776CMC@1: 89.43 -> 90.05Improved

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