https://github.com/masterskepticista/monocular-depth-estimation

Replicated results from DenseDepth using DenseNet169 in Python.
https://github.com/masterskepticista/monocular-depth-estimation

densenet-keras densenet169 monocular-depth monocular-depth-estimation monocular-slam nyu-depth-v2 nyuv2

Last synced: about 1 month ago
JSON representation

Replicated results from DenseDepth using DenseNet169 in Python.

• Host: GitHub
• URL: https://github.com/masterskepticista/monocular-depth-estimation
• Owner: MasterSkepticista
• Created: 2019-06-27T03:49:13.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2019-06-27T04:05:29.000Z (almost 7 years ago)
• Last Synced: 2025-03-06T10:42:49.400Z (about 1 year ago)
• Topics: densenet-keras, densenet169, monocular-depth, monocular-depth-estimation, monocular-slam, nyu-depth-v2, nyuv2
• Language: Python
• Size: 266 KB
• Stars: 1
• Watchers: 2
• Forks: 3
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Single Camera Depth Estimation using DenseNet169

Replicated results from DenseDepth using DenseNet169 in Python.

Ref: Original Work by Alhashim et al.

Run sketch.py to load data and start training.

Dataset: NYU-v2, more info can be found [here](https://cs.nyu.edu/~silberman/datasets/nyu_depth_v2.html)

Place the Dataset in the root directory. Download [here](https://s3-eu-west-1.amazonaws.com/densedepth/nyu_data.zip)

Sample Results: Trained on NVIDIA Tesla K80 (14GB VRAM); 3 epochs, bs 6, 4 hours

## Input 

![6 input images from indoor and outdoor](sample_in_snap.PNG)

## Output

![Output Depth Maps](sample_out_snap.PNG)

Notice that since the distribution of input dataset belongs to indoors, it performs reasonably well on indoors. 

```

@article{Alhashim2018,

  author    = {Ibraheem Alhashim and Peter Wonka},

  title     = {High Quality Monocular Depth Estimation via Transfer Learning},

  journal   = {arXiv e-prints},

  volume    = {abs/1812.11941},

  year      = {2018},

  url       = {https://arxiv.org/abs/1812.11941},

  eid       = {arXiv:1812.11941},

  eprint    = {1812.11941}

}

```