Cross domain Residual Transfer Learning for Person Re-identification - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2019

Cross domain Residual Transfer Learning for Person Re-identification


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.
Fichier principal
Vignette du fichier
furqan_wacv2019.pdf (1.98 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01947523 , version 1 (07-12-2018)


  • HAL Id : hal-01947523 , version 1


Furqan M. Khan, Francois F Bremond. Cross domain Residual Transfer Learning for Person Re-identification. WACV 2019 - IEEE’s and the PAMI-TC’s premier meeting on applications of computer vision, Jan 2019, Waikoloa Village, Hawaii, United States. ⟨hal-01947523⟩
66 View
268 Download


Gmail Facebook X LinkedIn More