Automatically Discovering Novel Visual Categories With Adaptive Prototype Learning
This article targets the task of novel category discovery (NCD), which aims to discover unknown categories when a certain number of classes are already known. The NCD task is challenging due to its closeness to real-world scenarios, where we have only encountered some partial classes and corresponding images. Unlike previous approaches to NCD, we propose a novel adaptive prototype learning method that leverages prototypes to emphasize category discrimination and alleviate the issue of missing annotations for novel classes. Concretely, the proposed method consists of two main stages: prototypical representation learning and prototypical self-training. In the first stage, we develop a robust feature extractor that could effectively handle images from both base and novel categories. This ability of instance and category discrimination of the feature extractor is boosted by self-supervised learning and adaptive prototypes. In the second stage, we utilize the prototypes again to rectify offline pseudo labels and train a final parametric classifier for category clustering. We conduct extensive experiments on four benchmark datasets, demonstrating our method's effectiveness and robustness with state-of-the-art performance.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2024 |
|---|---|
| Erschienen: |
2024 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:46 |
|---|---|
| Enthalten in: |
IEEE transactions on pattern analysis and machine intelligence - 46(2024), 4 vom: 23. März, Seite 2533-2544 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Zhang, Lu [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
|---|
| Anmerkungen: |
Date Revised 07.03.2024 published: Print-Electronic Citation Status PubMed-not-MEDLINE |
|---|
| doi: |
10.1109/TPAMI.2023.3335962 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM364864168 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM364864168 | ||
| 003 | DE-627 | ||
| 005 | 20240307232059.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2024 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1109/TPAMI.2023.3335962 |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1319.xml |
| 035 | |a (DE-627)NLM364864168 | ||
| 035 | |a (NLM)37995157 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Zhang, Lu |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Automatically Discovering Novel Visual Categories With Adaptive Prototype Learning |
| 264 | 1 | |c 2024 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Revised 07.03.2024 | ||
| 500 | |a published: Print-Electronic | ||
| 500 | |a Citation Status PubMed-not-MEDLINE | ||
| 520 | |a This article targets the task of novel category discovery (NCD), which aims to discover unknown categories when a certain number of classes are already known. The NCD task is challenging due to its closeness to real-world scenarios, where we have only encountered some partial classes and corresponding images. Unlike previous approaches to NCD, we propose a novel adaptive prototype learning method that leverages prototypes to emphasize category discrimination and alleviate the issue of missing annotations for novel classes. Concretely, the proposed method consists of two main stages: prototypical representation learning and prototypical self-training. In the first stage, we develop a robust feature extractor that could effectively handle images from both base and novel categories. This ability of instance and category discrimination of the feature extractor is boosted by self-supervised learning and adaptive prototypes. In the second stage, we utilize the prototypes again to rectify offline pseudo labels and train a final parametric classifier for category clustering. We conduct extensive experiments on four benchmark datasets, demonstrating our method's effectiveness and robustness with state-of-the-art performance | ||
| 650 | 4 | |a Journal Article | |
| 700 | 1 | |a Qi, Lu |e verfasserin |4 aut | |
| 700 | 1 | |a Yang, Xu |e verfasserin |4 aut | |
| 700 | 1 | |a Qiao, Hong |e verfasserin |4 aut | |
| 700 | 1 | |a Yang, Ming-Hsuan |e verfasserin |4 aut | |
| 700 | 1 | |a Liu, Zhiyong |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t IEEE transactions on pattern analysis and machine intelligence |d 1979 |g 46(2024), 4 vom: 23. März, Seite 2533-2544 |w (DE-627)NLM098212257 |x 1939-3539 |7 nnns |
| 773 | 1 | 8 | |g volume:46 |g year:2024 |g number:4 |g day:23 |g month:03 |g pages:2533-2544 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1109/TPAMI.2023.3335962 |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 46 |j 2024 |e 4 |b 23 |c 03 |h 2533-2544 | ||