CARLOR @ Ego4D 步骤定位挑战赛：用于测试时细化的贝叶斯时间顺序先验
CARLOR @ Ego4D Step Grounding Challenge: Bayesian temporal-order priors for test time refinement
AI总结：
针对 Ego4D 长时第一人称视频的步骤定位问题，提出 Bayesian-VSLNet，在推理时利用贝叶斯时间顺序先验处理循环重复动作并取得最先进性能。
摘要：
步骤定位任务的目标是根据自然语言描述确定活动的时间边界。本技术报告介绍了一种 Bayesian-VSLNet，以应对在冗长、未经裁剪的第一人称视频中识别此类时间段的挑战。我们的模型通过在推理过程中引入新颖的贝叶斯时间顺序先验，显著改进了传统模型，提高了时刻预测的准确性。该先验会针对视频中的循环和重复动作进行调整。我们的评估表明，其性能优于现有方法，在 Ego4D Goal-Step 数据集上取得了最先进结果：测试集上在 0.3 IoU 时 Recall Top-1 为 35.18，在 0.5 IoU 时 Recall Top-1 为 20.48。
英文摘要：
The goal of the Step Grounding task is to locate temporal boundaries of activities based on natural language descriptions. This technical report introduces a Bayesian-VSLNet to address the challenge of identifying such temporal segments in lengthy, untrimmed egocentric videos. Our model significantly improves upon traditional models by incorporating a novel Bayesian temporal-order prior during inference, enhancing the accuracy of moment predictions. This prior adjusts for cyclic and repetitive actions within videos. Our evaluations demonstrate superior performance over existing methods, achieving state-of-the-art results on the Ego4D Goal-Step dataset with a 35.18 Recall Top-1 at 0.3 IoU and 20.48 Recall Top-1 at 0.5 IoU on the test set.
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。
