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