Drift-Resilient Temporal Priors for Visual Tracking

TL;DR

This paper introduces DTPTrack, a module that enhances visual tracking by reducing model drift through temporal reliability calibration and guidance synthesis, improving performance across multiple architectures.

Contribution

The paper presents a novel, generalizable module that can be integrated into existing trackers to suppress drift and improve accuracy using calibrated temporal priors.

Findings

DTPTrack improves success rates on LaSOT and GOT-10k benchmarks.

Integration of DTPTrack yields consistent performance gains across diverse tracking architectures.

The best model achieves 77.5% Success on LaSOT and 80.3% AO on GOT-10k.

Abstract

Temporal information is crucial for visual tracking, but existing multi-frame trackers are vulnerable to model drift caused by naively aggregating noisy historical predictions. In this paper, we introduce DTPTrack, a lightweight and generalizable module designed to be seamlessly integrated into existing trackers to suppress drift. Our framework consists of two core components: (1) a Temporal Reliability Calibrator (TRC) mechanism that learns to assign a per-frame reliability score to historical states, filtering out noise while anchoring on the ground-truth template; and (2) a Temporal Guidance Synthesizer (TGS) module that synthesizes this calibrated history into a compact set of dynamic temporal priors to provide predictive guidance. To demonstrate its versatility, we integrate DTPTrack into three diverse tracking architectures--OSTrack, ODTrack, and LoRAT-and show consistent, significant performance gains across all baselines. Our best-performing model, built upon an extended LoRATv2 backbone, sets a new state-of-the-art on several benchmarks, achieving a 77.5% Success rate on LaSOT and an 80.3% AO on GOT-10k.