Self-Supervised RGB-T Tracking with Cross-Input Consistency

TL;DR

This paper introduces the first self-supervised RGB-T tracking method that learns from unlabeled video pairs, using cross-input consistency to achieve competitive performance without annotated data.

Contribution

It presents a novel self-supervised training strategy for RGB-T tracking based on cross-input consistency, eliminating the need for labeled RGB-T datasets.

Findings

Outperforms seven supervised RGB-T trackers on GTOT dataset

Effective training with only unlabeled RGB-T video pairs

Proves the viability of self-supervised learning in RGB-T tracking

Abstract

In this paper, we propose a self-supervised RGB-T tracking method. Different from existing deep RGB-T trackers that use a large number of annotated RGB-T image pairs for training, our RGB-T tracker is trained using unlabeled RGB-T video pairs in a self-supervised manner. We propose a novel cross-input consistency-based self-supervised training strategy based on the idea that tracking can be performed using different inputs. Specifically, we construct two distinct inputs using unlabeled RGB-T video pairs. We then track objects using these two inputs to generate results, based on which we construct our cross-input consistency loss. Meanwhile, we propose a reweighting strategy to make our loss function robust to low-quality training samples. We build our tracker on a Siamese correlation filter network. To the best of our knowledge, our tracker is the first self-supervised RGB-T tracker. Extensive experiments on two public RGB-T tracking benchmarks demonstrate that the proposed training strategy is effective. Remarkably, despite training only with a corpus of unlabeled RGB-T video pairs, our tracker outperforms seven supervised RGB-T trackers on the GTOT dataset.