JDAI-CV/CoTNet
CoTNet is a contextual transformer network that replaces standard convolutions with self-attention building blocks for visual recognition tasks.
Not currently ranked — collecting fresh signals.
star history
CoTNet is a unified self-attention building block that serves as an alternative to standard convolutions in ConvNets. The repository provides official PyTorch implementations of vision backbone models enhanced with contextualized self-attention for tasks including image classification, object detection, instance segmentation, and semantic segmentation. It achieves competitive accuracy with efficient inference time-accuracy trade-offs on ImageNet and MSCOCO benchmarks.
Frequently asked
- What is JDAI-CV/CoTNet?
- CoTNet is a contextual transformer network that replaces standard convolutions with self-attention building blocks for visual recognition tasks.
- Is CoTNet open source?
- Yes — JDAI-CV/CoTNet is an open-source project tracked on heatdrop.
- What language is CoTNet written in?
- JDAI-CV/CoTNet is primarily written in Python.
- How popular is CoTNet?
- JDAI-CV/CoTNet has 538 stars on GitHub.
- Where can I find CoTNet?
- JDAI-CV/CoTNet is on GitHub at https://github.com/JDAI-CV/CoTNet.