通过用于边界和预测选择的多样化推理策略增强弱监督视频定位
Enhancing Weakly Supervised Video Grounding via Diverse Inference Strategies for Boundary and Prediction Selection
AI总结：
本文针对弱监督视频定位中边界预测与top-1预测选择的问题，提出多样化推理策略，通过新颖的边界预测方法与考虑提议质量的选择方法，在无需额外训练的情况下有效提升了模型性能。
摘要：
弱监督视频定位旨在无需显式真实时间边界的情况下，定位与给定查询相关的时间边界。虽然现有方法主要使用基于高斯的提议，但它们忽略了推理过程中（1）边界预测和（2）top-1预测选择的重要性。在其边界预测中，边界仅简单设置在高斯均值两侧半个标准差处，这可能无法准确捕获最佳边界。在top-1预测过程中，这些现有方法严重依赖与其他提议的交集，而未考虑每个提议的不同质量。为解决这些问题，我们通过引入（1）新颖的边界预测方法以从多个高斯中捕获多样化边界，以及（2）考虑提议质量的新选择方法，探索了各种推理策略。在ActivityNet Captions和Charades-STA数据集上的大量实验验证了我们推理策略的有效性，证明了在无需额外训练的情况下实现了性能提升。
英文摘要：
Weakly supervised video grounding aims to localize temporal boundaries relevant to a given query without explicit ground-truth temporal boundaries. While existing methods primarily use Gaussian-based proposals, they overlook the importance of (1) boundary prediction and (2) top-1 prediction selection during inference. In their boundary prediction, boundaries are simply set at half a standard deviation away from a Gaussian mean on both sides, which may not accurately capture the optimal boundaries. In the top-1 prediction process, these existing methods rely heavily on intersections with other proposals, without considering the varying quality of each proposal. To address these issues, we explore various inference strategies by introducing (1) novel boundary prediction methods to capture diverse boundaries from multiple Gaussians and (2) new selection methods that take proposal quality into account. Extensive experiments on the ActivityNet Captions and Charades-STA datasets validate the effectiveness of our inference strategies, demonstrating performance improvements without requiring additional training.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
