Download
PyTorch training code and pretrained models for DETR (DEtection TRansformer). We replace the full complex hand-crafted object detection pipeline with a Transformer, and match Faster R-CNN with a ResNet-50, obtaining 42 AP on COCO using half the computation power (FLOPs) and the same number of parameters. Inference in 50 lines of PyTorch. What it is. Unlike traditional computer vision techniques, DETR approaches object detection as a direct set prediction problem. It consists of a set-based global loss, which forces unique predictions via bipartite matching, and a Transformer encoder-decoder architecture. Given a fixed small set of learned object queries, DETR reasons about the relations of the objects and the global image context to directly output the final set of predictions in parallel. Due to this parallel nature, DETR is very fast and efficient.
Features
- DETR is very simple to implement and experiment with
- We provide baseline DETR and DETR-DC5 models
- The models are also available via torch hub
- There are no extra compiled components in DETR and package dependencies are minimal
- We train DETR with AdamW setting learning rate in the transformer to 1e-4 and 1e-5 in the backbone
- We show that it is relatively straightforward to extend DETR to predict segmentation masks
Project Samples
