Seeing Objects in dark with Continual Contrastive Learning

TL;DR

This paper introduces a continual contrastive learning approach to enhance object detection robustness across varying conditions, especially at night, by aligning real and synthetic image representations while mitigating catastrophic forgetting.

Contribution

It proposes a novel contrastive learning method combined with continual learning to improve domain robustness in object detection, particularly under challenging night-time conditions.

Findings

Outperforms existing methods in night-time object detection scenarios

Effectively aligns real and synthetic image features to improve domain generalization

Reduces catastrophic forgetting during continual learning process

Abstract

Object Detection, a fundamental computer vision problem, has paramount importance in smart camera systems. However, a truly reliable camera system could be achieved if and only if the underlying object detection component is robust enough across varying imaging conditions (or domains), for instance, different times of the day, adverse weather conditions, etc. In an effort to achieving a reliable camera system, in this paper, we make an attempt to train such a robust detector. Unfortunately, to build a well-performing detector across varying imaging conditions, one would require labeled training images (often in large numbers) from a plethora of corner cases. As manually obtaining such a large labeled dataset may be infeasible, we suggest using synthetic images, to mimic different training image domains. We propose a novel, contrastive learning method to align the latent representations of a pair of real and synthetic images, to make the detector robust to the different domains. However, we found that merely contrasting the embeddings may lead to catastrophic forgetting of the information essential for object detection. Hence, we employ a continual learning based penalty, to alleviate the issue of forgetting, while contrasting the representations. We showcase that our proposed method outperforms a wide range of alternatives to address the extremely challenging, yet under-studied scenario of object detection at night-time.