恭喜我室的钟准博士的‘"Camera Style Adaptation for Person Re-identiﬁcation’被IEEE Conference on Computer Vision and Pattern Recognition, 2018录用。
Being a cross-camera retrieval task, person reidentiﬁcation suffers from image style variations caused by different cameras. The art implicitly addresses this problem by learning a camera-invariant descriptor subspace. In this paper, we explicitly consider this challenge by introducing camera style (CamStyle) adaptation. CamStyle can serve as a data augmentation approach that smooths the camera style disparities. Speciﬁcally, with CycleGAN, labeled training images can be style-transferred to each camera, and, along with the original training samples, form the augmented training set. This method, while increasing data diversity against over-ﬁtting, also incurs a considerable level of noise. In the effort to alleviate the impact of noise, the label smooth regularization (LSR) is adopted. The vanilla version of our method (without LSR) performs reasonably well on few-camera systems in which over-ﬁtting often occurs. With LSR, we demonstrate consistent improvement in all systems regardless of the extent of over-ﬁtting. We also report competitive accuracy compared with the state of the art.