Cross domain Residual Transfer Learning for Person Re-identification

Furqan Khan 1 Francois Bremond 1
1 STARS - Spatio-Temporal Activity Recognition Systems
CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : This paper presents a novel way to transfer model weights from one domain to another using residual learning framework instead of direct fine-tuning. It also argues for hybrid models that use learned (deep) features and statistical metric learning for multi-shot person re-identification when training sets are small. This is in contrast to popular end-to-end neural network based models or models that use hand-crafted features with adaptive matching models (neural nets or statistical metrics). Our experiments demonstrate that a hybrid model with residual transfer learning can yield significantly better re-identification performance than an end-to-end model when training set is small. On iLIDS-VID [42] and PRID [15] datasets, we achieve rank-1 recognition rates of 89.8% and 95%, respectively, which is a significant improvement over state-of-the-art.
Document type :
Conference papers
Complete list of metadatas
Contributor : Abhijit Das <>
Submitted on : Friday, December 7, 2018 - 3:11:11 PM
Last modification on : Thursday, February 28, 2019 - 10:48:01 AM
Long-term archiving on : Friday, March 8, 2019 - 1:02:40 PM


Files produced by the author(s)


  • HAL Id : hal-01947523, version 1



Furqan Khan, Francois Bremond. Cross domain Residual Transfer Learning for Person Re-identification. WACV 2019, Jan 2019, Waikoloa Village, Hawaii, United States. ⟨hal-01947523⟩



Record views


Files downloads