youngsikshin / image_pipeline Goto Github PK

Fork of image_pipeline

This package fills the gap between getting raw images from a camera driver and higher-level vision processing.

This Fork adds Tegra and fisheye support:

• camera_calibration_fisheye(https://github.com/DavidTorresOcana/image_pipeline/tree/indigo/camera_calibration_fisheye): A calibration tool for fisheye cameras. It uses OpenCV >3, NO CUDA support needed.
• image_proc_tegra(https://github.com/DavidTorresOcana/image_pipeline/tree/indigo/image_proc_tegra):

• A camera rectification package for camera rectification with OpenCV >3 and CUDA support. Tested in Tegra cores. Faster than image_proc since it uses CUDA for rectification

• It also implements a Nodelet for Fisheye cameras rectification, if calibrated with camera_calibration_fisheye(https://github.com/DavidTorresOcana/image_pipeline/tree/indigo/camera_calibration_fisheye). It uses OpenCV >3 with CUDA suport

• image_proc_fisheye (https://github.com/DavidTorresOcana/image_pipeline/tree/indigo/image_proc_fisheye): Rectification package for platforms without CUDA. Same as image_proc(https://github.com/DavidTorresOcana/image_pipeline/tree/indigo/image_proc) but for fisheye cameras calibrated with camera_calibration_fisheye(https://github.com/DavidTorresOcana/image_pipeline/tree/indigo/camera_calibration_fisheye). It uses OpenCV >3. NO CUDA support needed