arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.08918cs.CV
收藏

用于视频目标检测的运动矢量外推法

Motion Vector Extrapolation for Video Object Detection

Ryerson University(瑞尔森大学)

Julian True, Naimul Khan

摘要:

尽管用于视频目标检测的计算高效的深度神经网络架构不断取得成功,但性能不断达到速度与准确性与计算资源的巨大三元悖论(选择两个)。当前利用视频数据中的时间信息来克服这一三难困境的尝试受到目标检测模型中的最新技术的检验。我们提出了一种技术,它通过使用现成的对象检测器并行现有的基于光流的运动估计技术,执行视频目标检测。通过在基准MOT20数据集上的一组实验,我们证明了我们的方法在不牺牲任何准确性的情况下显著降低了任何给定对象检测器的基线延迟。进一步减少延迟,比原始延迟低25倍,可以实现最小的准确性损失。MOVEX可在基于CPU的常见系统上实现低延迟视频目标检测,从而实现GPU计算领域之外的高性能视频目标检测。该代码可在https://github.com/juliantrue/movex上获得。

英文摘要:

Despite the continued successes of computationally efficient deep neural network architectures for video object detection, performance continually arrives at the great trilemma of speed versus accuracy versus computational resources (pick two). Current attempts to exploit temporal information in video data to overcome this trilemma are bottlenecked by the state-of-the-art in object detection models. We present, a technique which performs video object detection through the use of off-the-shelf object detectors alongside existing optical flow based motion estimation techniques in parallel. Through a set of experiments on the benchmark MOT20 dataset, we demonstrate that our approach significantly reduces the baseline latency of any given object detector without sacrificing any accuracy. Further latency reduction, up to 25x lower than the original latency, can be achieved with minimal accuracy loss. MOVEX enables low latency video object detection on common CPU based systems, thus allowing for high performance video object detection beyond the domain of GPU computing. The code is available at https://github.com/juliantrue/movex.

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。

↑