Streaming video clips with large-scale video tokens impede vision transformers (ViTs) for efficient recognition, especially in video action detection where sufficient spatiotemporal representations are required for precise actor identification. In this work, we propose an end-to-end framework for efficient video action detection (EVAD) based on vanilla ViTs. Our EVAD consists of two specialized designs for video action detection. First, we propose a spatiotemporal token dropout from a keyframe-centric perspective. In a video clip, we maintain all tokens from its keyframe, preserve tokens relevant to actor motions from other frames, and drop out the remaining tokens in this clip. Second, we refine scene context by leveraging remaining tokens for better recognizing actor identities. The region of interest (RoI) in our action detector is expanded into temporal domain. The captured spatiotemporal actor identity representations are refined via scene context in a decoder with the attention mechanism. These two designs make our EVAD efficient while maintaining accuracy, which is validated on three benchmark datasets (i.e., AVA, UCF101-24, JHMDB). Compared to the vanilla ViT backbone, our EVAD reduces the overall GFLOPs by 43% and improves real-time inference speed by 40% with no performance degradation. Moreover, even at similar computational costs, our EVAD can improve the performance by 1.1 mAP with higher resolution inputs. Code is available at https://github.com/MCG-NJU/EVAD.

翻译：大规模视频令牌的流式视频片段阻碍了视觉Transformer（ViTs）的高效识别，特别是在需要充分的时空表征以实现精确行为者识别的视频动作检测任务中。本文提出了一种基于原始ViTs的端到端高效视频动作检测（EVAD）框架。我们的EVAD包含两项针对视频动作检测的专门设计。首先，我们从关键帧中心视角出发，提出了一种时空令牌丢弃方法。在视频片段中，我们保留关键帧的所有令牌，保留其他帧中与行为者运动相关的令牌，并丢弃该片段中的其余令牌。其次，我们通过利用剩余令牌来细化场景上下文，从而更好地识别行为者身份。动作检测器中的感兴趣区域（RoI）被扩展至时间域。捕捉到的时空行为者身份表征通过解码器中的注意力机制，借助场景上下文进行细化。这两项设计使我们的EVAD在保持准确性的同时实现高效，并在三个基准数据集（即AVA、UCF101-24、JHMDB）上得到验证。与原始ViT骨干网络相比，我们的EVAD在无性能损失的情况下，将整体GFLOPs降低了43%，实时推理速度提升了40%。此外，即使在相似计算成本下，我们的EVAD通过更高分辨率的输入也能将性能提升1.1 mAP。代码已开源在https://github.com/MCG-NJU/EVAD。