{
  "id": "2103.04778",
  "version": "v1",
  "published": "2021-03-08T14:16:09.000Z",
  "updated": "2021-03-08T14:16:09.000Z",
  "title": "Bridging the Distribution Gap of Visible-Infrared Person Re-identification with Modality Batch Normalization",
  "authors": [
    "Wenkang Li",
    "Qi Ke",
    "Wenbin Chen",
    "Yicong Zhou"
  ],
  "categories": [
    "cs.CV"
  ],
  "abstract": "Visible-infrared cross-modality person re-identification (VI-ReID), whose aim is to match person images between visible and infrared modality, is a challenging cross-modality image retrieval task. Most existing works integrate batch normalization layers into their neural network, but we found out that batch normalization layers would lead to two types of distribution gap: 1) inter-mini-batch distribution gap -- the distribution gap of the same modality between each mini-batch; 2) intra-mini-batch modality distribution gap -- the distribution gap of different modality within the same mini-batch. To address these problems, we propose a new batch normalization layer called Modality Batch Normalization (MBN), which normalizes each modality sub-mini-batch respectively instead of the whole mini-batch, and can reduce these distribution gap significantly. Extensive experiments show that our MBN is able to boost the performance of VI-ReID models, even with different datasets, backbones and losses.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2021-03-08T14:16:09.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "distribution gap",
      "modality batch normalization",
      "visible-infrared person re-identification",
      "cross-modality image retrieval task",
      "works integrate batch normalization layers"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}