Unsupervised Pre-Training for Detection Transformers

DEtection TRansformer (DETR) for object detection reaches competitive performance compared with Faster R-CNN via a transformer encoder-decoder architecture. However, trained with scratch transformers, DETR needs large-scale training data and an extreme long training schedule even on COCO dataset. In...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on pattern analysis and machine intelligence Vol. 45; no. 11; pp. 1 - 11
Main Authors: Dai, Zhigang, Cai, Bolun, Lin, Yugeng, Chen, Junying
Format: Journal Article
Language:English
Published: New York IEEE 01.11.2023
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Coders
Decoding
Encoders-Decoders
Feature extraction
Freezing
Image segmentation
Localization
Location awareness
Natural language processing
Object detection
Object recognition
one-shot detection
Queries
self-supervised learning
Task analysis
Training
transformer
Transformers
unsupervised pre-training
ISSN:0162-8828, 1939-3539, 2160-9292, 1939-3539
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:DEtection TRansformer (DETR) for object detection reaches competitive performance compared with Faster R-CNN via a transformer encoder-decoder architecture. However, trained with scratch transformers, DETR needs large-scale training data and an extreme long training schedule even on COCO dataset. Inspired by the great success of pre-training transformers in natural language processing, we propose a novel pretext task named random query patch detection in Unsupervised Pre-training DETR (UP-DETR). Specifically, we randomly crop patches from the given image and then feed them as queries to the decoder. The model is pre-trained to detect these query patches from the input image. During the pre-training, we address two critical issues: multi-task learning and multi-query localization. (1) To trade off classification and localization preferences in the pretext task, we find that freezing the CNN backbone is the prerequisite for the success of pre-training transformers. (2) To perform multi-query localization, we develop UP-DETR with multi-query patch detection with attention mask. Besides, UP-DETR also provides a unified perspective for fine-tuning object detection and one-shot detection tasks. In our experiments, UP-DETR significantly boosts the performance of DETR with faster convergence and higher average precision on object detection, one-shot detection and panoptic segmentation. Code and pre-training models: https://github.com/dddzg/up-detr .
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0162-8828
1939-3539
2160-9292
1939-3539
DOI:10.1109/TPAMI.2022.3216514