The project uses this KITTI dataset
for validation. First you will have to to place the downloaded set in the data/
directory, achieving the following directory structure:
data/
├── calib/
│ ├── 00
│ ├── 01
│ └── ...
├── poses/
│ ├── 00.txt
│ ├── 01.txt
│ └── ...
└── sequences/
├── 00
├── 01
└── ...
The directories can also be symlinks (that is encouraged actually).
Once you have set everything up properly, you can run the validation by running
./main.py. To see which options you have available, run
./main.py -h
- Anes Hadžić
- Haris Gušić