Product library

home page find GPTS Ranking List Build AI Submit market
home page Depth estimation
Depth estimation
Updated: 2024-06-14 17:20:36
newest Hot

Depth Anything V2

Advanced monocular depth estimation model

Depth Anything V2 is an improved monocular depth estimation model, which provides more precise and robust depth prediction than the previous version by using synthetic images and a large number of unlabeled real images for training. This model has significantly improved in efficiency and accuracy, and is more than 10 times faster than the latest model based on Stable Diffusion.
Depth estimation Monocular vision

Depth Anything

Unleashing the power of large-scale unlabeled data

Depth Anything is a highly practical solution for robust monocular depth estimation. We aim to build a simple and powerful basic model to process any image under any circumstances, without pursuing novel technical modules. To this end, we design a data engine to expand the data set, collect and automatically annotate large-scale unlabeled data (about 62M), so as to significantly expand the data coverage and reduce generalization errors. We have studied two simple and effective strategies to make data expansion promising. First, create more challenging optimization goals by leveraging data enhancement tools. It forces the model to actively seek additional visual knowledge and obtain strong representation. Secondly, an auxiliary supervision is developed to force the model to inherit rich semantic prior from the pre training encoder. We conducted an extensive evaluation of its zero shot capability, including six public data sets and random photos. It shows impressive generalization. In addition, by using the measurement depth information from NYUv2 and KITTI to fine tune them, we have established new SOTAs. Our better depth model also leads to better depth condition ControlNet. Our model is published in https://github.com/LiheYoung/Depth-Anything 。
Selected Chinese
Depth estimation image processing

Dpt Depth

Dpt depth estimation+3D

Dpt Depth is an image processing tool based on Dpt depth estimation and 3D technology. It can quickly estimate the depth information from the input image, and generate the corresponding 3D model according to the depth information. Dpt Depth Estimation+3D is powerful and easy to use, and can be widely used in computer vision, image processing and other fields. The product provides free trial version and paid subscription version.
Depth estimation 3D

Control-LoRA

Model control technology based on low rank parameter optimization

Control LoRA is implemented by adding low rank parameter optimization to ControlNet, which provides a more efficient and compact model control method for consumer grade GPUs. This product includes multiple Control LoRA models, including MiDaS and ClipDrop depth estimation, Canny edge detection, photo and sketch coloring, revision and other functions. After training, the Control LoRA model can generate high-quality images on different image concepts and aspect ratios.
image processing model control

© 2024     AIbase    Record No.: Min ICP Bei 08105208 No.14

Privacy Policy

User Agreement

Feedback Site Map