veri-wild

A pytorch codebase for human parsing and vehicle parsing

This repo gives the code for the paper "Xinchen Liu, Wu Liu, Jinkai Zheng, Chenggang Yan, Tao Mei: Beyond the Parts: Learning Multi-view Cross-part Correlation for Vehicle Re-identification. ACM MM 2020".

Code for Part-Guided Attention Learning for Vehicle Instance Retrieval (TITS2020). A strong Vehicle Re-ID model with part region guidance.

Unsupervised clustering of 400K+ vehicle images using CNN descriptors (EfficientNet-B7, OSNet, VDC). Compares preprocessing, clustering algorithms, and DBSCAN outlier detection for an IntelliVision Smart City case study.