SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2102.08318 · CVPR · 2021

Instance Localization for Self-supervised Detection Pretraining

Zhirong Wu, Stephen Lin, Bolei Zhou, Ceyuan Yang

arXiv · PDF · Open in the Atlas

Code that ran

We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Prior research on self-supervised learning has led to considerable progress on image classification, but often with degraded transfer performance on object detection. The objective of this paper is to advance self-supervised pretrained models specifically for object detection. Based on the inherent difference between classification and detection, we propose a new self-supervised pretext task, called instance localization. Image instances are pasted at various locations and scales onto background images. The pretext task is to predict the instance category given the composited images as well as the foreground bounding boxes. We show that integration of bounding boxes into pretraining promotes better task alignment and architecture alignment for transfer learning. In addition, we propose an augmentation method on the bounding boxes to further enhance the feature alignment. As a result, our model becomes weaker at Imagenet semantic classification but stronger at image patch localization, with an overall stronger pretrained model for object detection. Experimental results demonstrate that our approach yields state-of-the-art transfer learning results for object detection on PASCAL VOC and MSCOCO 1 .

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2102.08318")
get_code_for_paper("2102.08318")
have("2102.08318")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2102.08318.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2102.08318.svg)](https://syntology.ai/paper/2102.08318)

Connect an agent — have() is free.