ENACT：基于熵的注意力输入聚类以降低目标检测Transformers的计算需求
ENACT: Entropy-based Clustering of Attention Input for Reducing the Computational Needs of Object Detection Transformers
AI总结：
提出基于熵聚类的ENACT模块作为多头自注意力Transformer的插件，通过减少GPU内存使用同时保持合理精度，解决目标检测Transformer计算资源需求大的问题。
摘要：
Transformers在基于视觉的目标检测问题上展现出极具竞争力的精度。然而，由于注意力权重的二次方规模，它们需要大量的计算资源。在本研究中，由于同一目标的像素之间存在相似性，我们提出基于其熵对transformer输入进行聚类。这有望在保持合理精度的同时，减少训练期间的GPU使用量。这一想法通过一个名为ENACT（基于熵的注意力聚类检测Transformers）的已实现模块得以实现，该模块可作为插件用于任何基于多头自注意力机制的transformer网络。在COCO目标检测数据集和三种检测transformers上的实验表明，内存需求得到降低，而检测精度仅略有下降。ENACT模块的代码可在https://github.com/GSavathrakis/ENACT获取。
英文摘要：
Transformers demonstrate competitive performance in terms of precision on the problem of vision-based object detection. However, they require considerable computational resources due to the quadratic size of the attention weights. In this work, we propose to cluster the transformer input on the basis of its entropy, due to its similarity between same object pixels. This is expected to reduce GPU usage during training, while maintaining reasonable accuracy. This idea is realized with an implemented module that is called ENtropy-based Attention Clustering for detection Transformers (ENACT), which serves as a plug-in to any multi-head self-attention based transformer network. Experiments on the COCO object detection dataset and three detection transformers demonstrate that the requirements on memory are reduced, while the detection accuracy is degraded only slightly. The code of the ENACT module is available at https://github.com/GSavathrakis/ENACT.
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。
