RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning. [ICLR 2026]

computer-vision detr instance-segmentation machine-learning object-detection rf-detr sota
4 Open Issues Need Help Last updated: Jul 5, 2026

Open Issues Need Help

View All on GitHub
enhancement good first issue

RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning. [ICLR 2026]

Python
#computer-vision#detr#instance-segmentation#machine-learning#object-detection#rf-detr#sota
enhancement help wanted

RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning. [ICLR 2026]

Python
#computer-vision#detr#instance-segmentation#machine-learning#object-detection#rf-detr#sota
enhancement good first issue

RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning. [ICLR 2026]

Python
#computer-vision#detr#instance-segmentation#machine-learning#object-detection#rf-detr#sota

AI Summary: Custom dataset classes are not being read when their `supercategory` field is set to `none`, leading to an empty classes tensor and a training crash. The user has identified a specific line in `rfdetr/detr.py` as the potential source of this bug.

Complexity: 2/5
bug good first issue

RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning. [ICLR 2026]

Python
#computer-vision#detr#instance-segmentation#machine-learning#object-detection#rf-detr#sota