Corpus ID: 211031851

Illumination adaptive person reid based on teacher-student model and adversarial training

@article{Zhang2020IlluminationAP,
  title={Illumination adaptive person reid based on teacher-student model and adversarial training},
  author={Ziyue Zhang and Richard Y. D. Xu and Shuai Jiang and Yang Li and Congzhentao Huang and Chen Deng},
  journal={ArXiv},
  year={2020},
  volume={abs/2002.01625}
}
  • Ziyue Zhang, Richard Y. D. Xu, +3 authors Chen Deng
  • Published in ArXiv 2020
  • Computer Science
  • Most existing works in Person Re-identification (ReID) focus on settings where illumination either is kept the same or has very little fluctuation. However, the changes in the illumination degree may affect the robustness of a ReID algorithm significantly. To address this problem, we proposed a Two-Stream Network that can separate ReID features from lighting features to enhance ReID performance. Its innovations are threefold: (1) A discriminative entropy loss to ensure the ReID features contain… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 22 REFERENCES

    Scalable person reidentification: A benchmark

    • Liang Zheng, Liyue Shen, +3 authors Qi Tian
    • ICCV, 2015, pp. 1116– 1124.
    • 2015
    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Bag of Tricks and a Strong Baseline for Deep Person Re-Identification

    Joint Discriminative and Generative Learning for Person Re-Identification

    VIEW 2 EXCERPTS

    Omni-Scale Feature Learning for Person Re-Identification

    Progressive Pose Attention Transfer for Person Image Generation

    VIEW 1 EXCERPT

    Vrstc: Occlusionfree video person re-identification

    • Ruibing Hou, Bingpeng Ma, +3 authors Xilin Chen
    • CVPR, 2019, pp. 7183–7192.
    • 2019
    VIEW 1 EXCERPT

    Adversarially Occluded Samples for Person Re-identification

    VIEW 1 EXCERPT

    Exploring Disentangled Feature Representation Beyond Face Identification

    VIEW 1 EXCERPT

    Harmonious Attention Network for Person Re-identification