Depth to Disparity Converter

Convert EXR depth images to TIFF disparity images for use with the Nodar Viewer.

Installation

pip install -e examples/python/depth_to_disparity

Usage

cd examples/python/depth_to_disparity/depth_to_disparity
OPENCV_IO_ENABLE_OPENEXR=1 python3 depth_to_disparity <data_directory> [output_directory]

Parameters

  • data_directory: Path to directory containing depth and details folders from Hammerhead
  • output_directory: Optional output directory (defaults to disparity folder in data_directory)

Examples

# Convert depth images with default output location
OPENCV_IO_ENABLE_OPENEXR=1 python3 depth_to_disparity /path/to/hammerhead/data

# Convert depth images to specific output directory
OPENCV_IO_ENABLE_OPENEXR=1 python3 depth_to_disparity /path/to/hammerhead/data /path/to/output

Output

  • Format: TIFF disparity images (.tiff)
  • Location: disparity folder in data directory (or specified output directory)
  • Naming: Maintains original depth image naming convention

Features

  • Convert EXR depth images to lossless TIFF disparity format
  • Compatible with Nodar Viewer for point cloud generation
  • Preserves depth information accuracy
  • Batch processing of entire directories

Requirements

  • OpenCV with EXR support enabled
  • Both depth and details folders in input directory
  • Details data must be in YAML format

Troubleshooting

  • EXR support missing: Install OpenCV with EXR support or use OPENCV_IO_ENABLE_OPENEXR=1
  • Missing details folder: Ensure both depth and details folders exist in data directory
  • File overwrite warning: Existing files in output directory will be overwritten
  • Orin compatibility: Default OpenCV on Orin may lack EXR support - use x86-64 system

Press Ctrl+C to stop conversion.