search

JDAI-CV / VeRidataset

This is the project page for veri dataset which is a large scale image dataset for vehicle re-identification in urban traffic surveillance.
399 stars 55 forks source link

ICCV 2019 paper to be added #10

Closed zhengthomastang closed 5 years ago

zhengthomastang commented 5 years ago

We have a newly published ICCV 2019 paper with experiments on VeRi-776. Hope it could be added to the state-of-the-art results and references.

@inproceedings{Tang19PAMTRI, author = {Zheng Tang and Milind Naphade and Stan Birchfield and Jonathan Tremblay and William Hodge and Ratnesh Kumar and Shuo Wang and Xiaodong Yang}, title = {{PAMTRI}: {P}ose-aware multi-task learning for vehicle re-identification using highly randomized synthetic data}, booktitle = {Proc. ICCV}, pages = {211--220}, address = {Seoul, Korea}, year = {2019} }

Link: http://openaccess.thecvf.com/content_ICCV_2019/papers/Tang_PAMTRI_Pose-Aware_Multi-Task_Learning_for_Vehicle_Re-Identification_Using_Highly_Randomized_ICCV_2019_paper.pdf

Our state-of-the-art performance is as follow (presented in Table 2 of the paper): mAP: 71.88% Rank-1: 92.86% Rank-5: 96.97%

taosean commented 5 years ago

Hi @zhengthomastang , I noticed there is another paper claims to achieve state-of-the-art performance on VeRi dataset, they claim to achieve the performance: (in Table III) mAP: 72.5 Rank-1: 93.3 Rank-5: 97.1 How do you think about this?

The paper is: Stripe-based and Attribute-aware Network: A Two-Branch Deep Model for Vehicle Re-identification.

zhengthomastang commented 5 years ago

@taosean Thank you for sharing! Their performance is really impressive. It would be great if both our and their papers could be listed among the state-of-the-art on the benchmark.

VehicleReId commented 5 years ago

Recent results have been updated. Thanks!