{
  "id": "2302.01512",
  "version": "v1",
  "published": "2023-02-03T02:57:18.000Z",
  "updated": "2023-02-03T02:57:18.000Z",
  "title": "Spectral Aware Softmax for Visible-Infrared Person Re-Identification",
  "authors": [
    "Lei Tan",
    "Pingyang Dai",
    "Qixiang Ye",
    "Mingliang Xu",
    "Yongjian Wu",
    "Rongrong Ji"
  ],
  "categories": [
    "cs.CV"
  ],
  "abstract": "Visible-infrared person re-identification (VI-ReID) aims to match specific pedestrian images from different modalities. Although suffering an extra modality discrepancy, existing methods still follow the softmax loss training paradigm, which is widely used in single-modality classification tasks. The softmax loss lacks an explicit penalty for the apparent modality gap, which adversely limits the performance upper bound of the VI-ReID task. In this paper, we propose the spectral-aware softmax (SA-Softmax) loss, which can fully explore the embedding space with the modality information and has clear interpretability. Specifically, SA-Softmax loss utilizes an asynchronous optimization strategy based on the modality prototype instead of the synchronous optimization based on the identity prototype in the original softmax loss. To encourage a high overlapping between two modalities, SA-Softmax optimizes each sample by the prototype from another spectrum. Based on the observation and analysis of SA-Softmax, we modify the SA-Softmax with the Feature Mask and Absolute-Similarity Term to alleviate the ambiguous optimization during model training. Extensive experimental evaluations conducted on RegDB and SYSU-MM01 demonstrate the superior performance of the SA-Softmax over the state-of-the-art methods in such a cross-modality condition.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2023-02-03T02:57:18.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "visible-infrared person re-identification",
      "spectral aware softmax",
      "match specific pedestrian images",
      "softmax loss training paradigm",
      "original softmax loss"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}