用验证模型筛选伪标签，让追踪模型在真实视频中更准更省数据。
Real-World Point Tracking with Verifier-Guided Pseudo-Labeling
- 引入验证器模型评估多个追踪器预测的可信度，筛选高质量伪标签。
- 在四个真实场景数据集上实现当前最佳性能，且所需标注数据更少。
- 适合需要低成本适配真实视频的点追踪应用，如自动驾驶、视频分析。
长期点追踪模型通常在大规模合成数据集上训练，但在真实视频中因特性差异和缺乏密集真值标注而性能下降。自训练在无标签视频上被探索作为实用方案，但伪标签质量高度依赖教师模型的可靠性，而该可靠性随帧和场景变化。本文提出验证器（verifier）——一个元模型，可学习评估追踪器预测的可靠性并指导伪标签生成。给定多个预训练追踪器的候选轨迹，验证器逐帧评估并选取最可信预测，生成高质量伪标签轨迹。应用于微调时，验证器引导的伪标签显著提升监督信号质量，实现对无标签视频的数据高效适应。在四个真实世界基准上的大量实验表明，本方法达到当前最优效果，且所需数据量少于先前自训练方法。
原文摘要 · Abstract (English)
Models for long-term point tracking are typically trained on large synthetic datasets. The performance of these models degrades in real-world videos due to different characteristics and the absence of dense ground-truth annotations. Self-training on unlabeled videos has been explored as a practical solution, but the quality of pseudo-labels strongly depends on the reliability of teacher models, which vary across frames and scenes. In this paper, we address the problem of real-world fine-tuning and introduce verifier, a meta-model that learns to assess the reliability of tracker predictions and guide pseudo-label generation. Given candidate trajectories from multiple pretrained trackers, the verifier evaluates them per frame and selects the most trustworthy predictions, resulting in high-quality pseudo-label trajectories. When applied for fine-tuning, verifier-guided pseudo-labeling substantially improves the quality of supervision and enables data-efficient adaptation to unlabeled videos. Extensive experiments on four real-world benchmarks demonstrate that our approach achieves state-of-the-art results while requiring less data than prior self-training methods. Project page: https://kuis-ai.github.io/track_on_r
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。