OccCasNet: Occlusion-aware Cascade Cost Volume for Light Field Depth Estimation

Network Architecture

SOTA on 4D Light Field Benchmark

• Our method achieve competitive performance on the HCI 4D LF Benchmark in terms of all the five accuracy metrics (i.e., BadPix0.01, BadPix0.03, BadPix0.07, MSE and Q25).

• For more detail comparison, please use the link below.
• Benchmark link

Environment

Ubuntu            16.04
Python            3.8.10
Tensorflow-gpu    2.5.0
CUDA              11.2

Train OccCasNet

1. Download HCI Light field dataset from http://hci-lightfield.iwr.uni-heidelberg.de/.
2. Unzip the LF dataset and move 'additional/, training/, test/, stratified/ ' into the 'hci_dataset/'.
3. Stage 1: Run python train_occcas.py
   • Checkpoint files will be saved in 'LF_checkpoints/XXX_ckp/iterXXXX_valmseXXXX_bpXXX.hdf5'.
   • Training process will be saved in
   • 'LF_output/XXX_ckp/train_iterXXXXX.jpg'
   • 'LF_output/XXX_ckp/val_iterXXXXX.jpg'.

Evaluate OccCasNet

• Run python evaluation_occcas.py
  • path_weight='LF_checkpoint/SubFocal_sub_0.5_js_0.1_ckp/iter0010_valmse0.768_bp1.93.hdf5'

Submit OccCasNet

• Run python submission_occcas.py
  • path_weight='LF_checkpoint/SubFocal_sub_0.5_js_0.1_ckp/iter0010_valmse0.768_bp1.93.hdf5'

Last modified data: 2023/05/28.

The code is modified and heavily borrowed from LFattNet: https://github.com/LIAGM/LFattNet, SubFocal: https://github.com/chaowentao/SubFocal

The code they provided is greatly appreciated.