ghost233lism

/

DepthAnything-AC

like 18

This PR significantly improves the model card for DepthAnything-AC by incorporating more detailed information from the paper and its associated GitHub repository.

Key improvements include:

• Metadata: Added library_name: transformers tag for better discoverability and integration.
• Paper Abstract: Included the full abstract to provide a clear overview of the research.
• Project Page & Code: Added prominent links to the project page (https://ghost233lism.github.io/depthanything-AC-page/) and the GitHub repository (https://github.com/HVision-NKU/DepthAnythingAC).
• Comprehensive Introduction: Expanded the introduction section with details from the GitHub README.
• Model Architecture: Added a section to describe the model's architecture.
• Detailed Usage: Enhanced the usage section with instructions on model download, quick inference, training, and evaluation, mirroring the GitHub guide.
• Quantitative Results: Included the quantitative performance tables from the paper's GitHub README for various benchmarks.
• License, Contact, and Acknowledgements: Added explicit sections for the license details, contact information, and acknowledgements as found in the GitHub README, ensuring all relevant information is present on the Hugging Face Hub.

These updates aim to provide researchers with a more complete and accessible understanding of the model, its capabilities, and how to use it.