Depth Anything is a foundation model for monocular depth estimation trained on large-scale unlabeled data. The vitb14 variant uses a Vision Transformer backbone. It produces high-quality relative depth maps from RGB images and is available through the Transformers library and the official GitHub repository.
In the Other AI space, Depth Anything Vitb14 takes a focused approach. It focuses on estimating depth from a single image without specialized depth sensors. Depth Anything Vitb14 is an open-source project aimed at developers. The project is open source (Open Source). Depth Anything Vitb14 is available on the web, the command line, and API.
Behind Depth Anything Vitb14 is LiheYoung, and it first shipped in 2024.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do