ReFace: Improving Clothes-Changing
Re-Identification With Face Features

1Reichman University
2The Hebrew University of Jerusalem, Israel

*Equal Contribution

Models comparison on the 42Street dataset. The percentage next to the name of every model indicates the accuracy of the model.


Person re-identification (ReID) has been an active research field for many years. Despite that, models addressing this problem tend to perform poorly when the task is to re-identify the same people over a prolonged time, due to appearance changes such as different clothes and hairstyles. In this work, we introduce a new method that takes full advantage of the ability of existing ReID models to extract appearance-related features and combines it with a face feature extraction model to achieve new state-of-the-art results, both on image-based and video-based benchmarks. Moreover, we show how our method could be used for an application in which multiple people of interest, under clothes-changing settings, should be re-identified given an unseen video and a limited amount of labeled data. We claim that current ReID benchmarks do not represent such real-world scenarios, and publish a new dataset, 42Street, based on a theater play as an example of such an application. We show that our proposed method outperforms existing models also on this dataset while using only pre-trained modules and without any further training.

Visual Results On The 42Street Dataset

Closed Set Setting

All identities of the query samples have at least one corresponding image in the gallery.

Open Set Setting

A query identity may not appear in the gallery.

Method Main Idea - Gallery Enrichment

The gallery enrichment process. Given labeled data and unlabeled query samples, we enrich the gallery using query images that include faces.


  title={ReFace: Improving Clothes-Changing Re-Identification With Face Features},
  author={Arkushin, Daniel and Cohen, Bar and Peleg, Shmuel and Fried, Ohad},
  journal={arXiv preprint arXiv:2211.13807},