Sparse DETR: Efficient End-to-End Object Detection with Learnable Sparsity

Abstract

DETR is the first end-to-end object detector using a transformer encoder-decoder architecture and demonstrates competitive performance but low computational efficiency on high resolution feature maps. The subsequent work, Deformable DETR, enhances the efficiency of DETR by replacing dense attention with deformable attention, which achieves 10x faster convergence and improved performance. Deformable DETR uses the multiscale feature to ameliorate performance, however, the number of encoder tokens increases by 20x compared to DETR, and the computation cost of the encoder attention remains a bottleneck. In our preliminary experiment, we observe that the detection performance hardly deteriorates even if only a part of the encoder token is updated. Inspired by this observation, we propose Sparse DETR that selectively updates only the tokens expected to be referenced by the decoder, thus help the model effectively detect objects. In addition, we show that applying an auxiliary detection loss on the selected tokens in the encoder improves the performance while minimizing computational overhead. We validate that Sparse DETR achieves better performance than Deformable DETR even with only 10% encoder tokens on the COCO dataset. Albeit only the encoder tokens are sparsified, the total computation cost decreases by 38% and the frames per second (FPS) increases by 42% compared to Deformable DETR. Code is available at https://github.com/kakaobrain/sparse-detr

ICLR 2022. Code is available at https://github.com/kakaobrain/sparse-detr

References in corpus (10)

• Generating Long Sequences with Sparse Transformers
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• Axial Attention in Multidimensional Transformers
• Reformer: The Efficient Transformer
• DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification
• Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention
• Efficient DETR: Improving End-to-End Object Detector with Dense Prior
• IA-RED${}^2$: Interpretability-Aware Redundancy Reduction for Vision Transformers
• Spatially Consistent Representation Learning
• What Makes for End-to-End Object Detection?