Abstract
Unsupervised Deep Distance Metric Learning (UDML) aims to learn sample simi- larities in the embedding space from an unlabeled dataset. Traditional UDML methods usually use the triplet loss or pairwise loss which requires the mining of positive and neg- ative samples w.r.t. anchor data points. This is, however, challenging in an unsupervised setting as the label information is not available. In this paper, we propose a new UDML method that overcomes that challenge. In particular, we propose to use a deep cluster- ing loss to learn centroids, i.e., pseudo labels, that represent semantic classes. During learning, these centroids are also used to reconstruct the input samples. It hence ensures the representativeness of centroids — each centroid represents visually similar samples. Therefore, the centroids give information about positive (visually similar) and negative (visually dissimilar) samples. Based on pseudo labels, we propose a novel unsupervised metric loss which enforces the positive concentration and negative separation of samples in the embedding space. Experimental results on benchmarking datasets show that the proposed approach outperforms other UDML methods.
Original language | English |
---|---|
Title of host publication | 31st British Machine Vision Conference, BMVC 2020 |
Editors | Neill Campbell |
Place of Publication | London UK |
Publisher | British Machine Vision Association and Society for Pattern Recognition |
Number of pages | 13 |
Publication status | Published - 2020 |
Event | British Machine Vision Conference 2020 - Virtual, London, United Kingdom Duration: 7 Sep 2020 → 10 Sep 2020 Conference number: 31st https://www.bmvc2020-conference.com (Website) https://www.bmvc2020-conference.com/programme/accepted-papers/ (Proceedings) |
Conference
Conference | British Machine Vision Conference 2020 |
---|---|
Abbreviated title | BMVC 2020 |
Country | United Kingdom |
City | London |
Period | 7/09/20 → 10/09/20 |
Internet address |