OpenGVLab/InternImage
InternImage is a vision foundation model architecture using deformable convolutions that achieves state-of-the-art results on object detection and semantic segmentation benchmarks.
Not currently ranked — collecting fresh signals.
star history
InternImage is a large-scale vision foundation model that adapts deformable convolutions for scalable deep learning on visual tasks. It serves as a general-purpose backbone for detection and segmentation tasks including COCO object detection, LVIS, and Pascal VOC. The model was highlighted at CVPR 2023 and demonstrates competitive performance against transformer-based vision models on multiple benchmarks.
Frequently asked
- What is OpenGVLab/InternImage?
- InternImage is a vision foundation model architecture using deformable convolutions that achieves state-of-the-art results on object detection and semantic segmentation benchmarks.
- Is InternImage open source?
- Yes — OpenGVLab/InternImage is open source, released under the MIT license.
- What language is InternImage written in?
- OpenGVLab/InternImage is primarily written in Python.
- How popular is InternImage?
- OpenGVLab/InternImage has 2.8k stars on GitHub.
- Where can I find InternImage?
- OpenGVLab/InternImage is on GitHub at https://github.com/OpenGVLab/InternImage.