PyTorch implementation of I. Alhashim and P. Wonka, “High Quality Monocular Depth Estimation via Transfer Learning,” CoRR, vol. abs/1812.11941, [LINK]
To verify the correct implementation of the model, the model is trained on the NYU_V2 dataset as in the paper. The model is currently used to create a 3D map of a robot's surroundings, enabling it to detect and avoid obstacles.
pip install -r requirements.txt
python train.py --batch-size 8 --epochs 20 --learning-rate 0.0001