# Pair DETR: Contrastive Learning Speeds Up DETR Training

Source: [Pair DETR: Contrastive Learning Speeds Up DETR Training](<https://www.sota2.com/research/paper/68fa6de71e209131cea3ccfa>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T00:46:11.149Z

Authors: Seyed Mehdi Iranmanesh, Xiaotong Chen, Kuo-Chin Lien

Published: 2022-10-29T03:02:49.000Z

## Abstract

The DETR object detection approach applies the transformer encoder and decoder architecture to detect objects and achieves promising performance. In this paper, we present a simple approach to address the main problem of DETR, the slow convergence, by using representation learning technique. In this approach, we detect an object bounding box as a pair of keypoints, the top-left corner and the center, using two decoders. By detecting objects as paired keypoints, the model builds up a joint classification and pair association on the output queries from two decoders. For the pair association we propose utilizing contrastive self-supervised learning algorithm without requiring specialized architecture. Experimental results on MS COCO dataset show that Pair DETR can converge at least 10x faster than original DETR and 1.5x faster than Conditional DETR during training, while having consistently higher Average Precision scores.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2210.16476>)

[PDF](<https://arxiv.org/pdf/2210.16476.pdf>)