End-to-End Object Detection with Adaptive Clustering Transformer
This work provides an incremental improvement in computational efficiency for end-to-end object detection, primarily benefiting researchers and practitioners working with Transformer-based object detection models.
This paper proposes Adaptive Clustering Transformer (ACT) to address the high computational cost of DETR for high-resolution inputs. ACT reduces the quadratic complexity of self-attention from O(N^2) to O(NK) by adaptively clustering query features and approximating interactions, achieving a good balance between accuracy and computation.
End-to-end Object Detection with Transformer (DETR)proposes to perform object detection with Transformer and achieve comparable performance with two-stage object detection like Faster-RCNN. However, DETR needs huge computational resources for training and inference due to the high-resolution spatial input. In this paper, a novel variant of transformer named Adaptive Clustering Transformer(ACT) has been proposed to reduce the computation cost for high-resolution input. ACT cluster the query features adaptively using Locality Sensitive Hashing (LSH) and ap-proximate the query-key interaction using the prototype-key interaction. ACT can reduce the quadratic O(N2) complexity inside self-attention into O(NK) where K is the number of prototypes in each layer. ACT can be a drop-in module replacing the original self-attention module without any training. ACT achieves a good balance between accuracy and computation cost (FLOPs). The code is available as supplementary for the ease of experiment replication and verification. Code is released at \url{https://github.com/gaopengcuhk/SMCA-DETR/}