Bot releases are hidden (Show)
- New models based on 5 different types of transformers (BEiT, Swin2, Swin, Next-ViT, LeViT)
- Training datasets extended from 10 to 12, including also KITTI and NYU Depth V2 using BTS split
- Best model, BEiTLarge 512, with resolution 512x512, is on average about 28% more accurate than MiDaS v3.0
- Integrated live depth estimation from camera feed
MiDaS
- MiDaS v3 (DPT)
Published by thias15 almost 2 years ago
- New models based on Dense Prediction Transformers are on average 21% more accurate than MiDaS v2.1
- Additional DPT models can be found here
MiDaS
- MiDaS v2.1
Published by ranftlr almost 4 years ago
- New model that was trained on 10 datasets and is on average about 10% more accurate
- New light-weight model that achieves real-time performance on mobile platforms
- Sample applications for iOS and Android
- ROS package for easy deployment on robots
MiDaS
- MiDaS v2
Published by ranftlr over 4 years ago
MiDaS
- MiDaS v1
Published by ranftlr almost 5 years ago