For my master thesis I have to get depth estimation from a monocular images using neural networks. I am familiar with neural networks but I dont know how to get "Ground truth images" , and how can I get depth maps from a single monocular image. Basically I am searching for a start in these directions.
Any kind if help will be appreciated.
Thankyou!
You should start by reading relevant papers [1-6].
Next, once you have an idea about your approach, check for appropriate datasets, e.g., Cityscapes, Make3D, NYUDepth.
Good luck!
[1] J. Uhrig et al., Pixel-level Encoding and Depth Layering for Instance-level Semantic Labeling, 2016
[2] D. Eigen et al., Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecture, 2015
[3] B. Li et al., Depth and surface normal estimation from monocular images using regression on deep features and hierarchical CRFs, 2015
[4] F. Liu et al., Deep Convolutional Neural Fields for Depth Estimation from a Single Image, 2015
[5] D. Eigen et al., Depth Map Prediction from a Single Image using a Multi-Scale Deep Network, 2014
[6] L. Ladicky et al., Pulling Things Out of Perspective, 2014
Hey @ MarcoSeeland!
Thankyou for your reply . I started reading some papers but i got lost in between due to new concepts and terminology . I was hoping if I could get some aid in video types or something which is easy to read to develop my approach.
- Badges
- Science topic
- Similar topics
- Computer Science
- Images