Is there anyway to get full sensor field of view in the libcamera pipeline? All examples are Pi v2 camera.
Using libcamera and ffmpeg APIs I can get 720p final image that uses the (mostly) full field of view – set sensor to 1640x1232 and pass those frames through ffmpeg filter of crop (4, 157)/1632x918 and then scale to 1280x720.
As the libcamera::ScalerCrop control states “Cropping is controlled explicitly through the controls::ScalerCrop control, while scaling is controlled implicitly based on the crop rectangle and the output stream size. The crop rectangle is expressed relatively to the full pixel array size and indicates how the field of view is affected by the pipeline.” When setting output size to 1280x720 the max ScalerCrop goes to (680, 692)/1920x1080 (this implies that the mode selected is 1920x1080).
In short is there anyway to get the behavior if the combo libcamera and ffmpeg pipeline using just libcamera?
Using libcamera and ffmpeg APIs I can get 720p final image that uses the (mostly) full field of view – set sensor to 1640x1232 and pass those frames through ffmpeg filter of crop (4, 157)/1632x918 and then scale to 1280x720.
As the libcamera::ScalerCrop control states “Cropping is controlled explicitly through the controls::ScalerCrop control, while scaling is controlled implicitly based on the crop rectangle and the output stream size. The crop rectangle is expressed relatively to the full pixel array size and indicates how the field of view is affected by the pipeline.” When setting output size to 1280x720 the max ScalerCrop goes to (680, 692)/1920x1080 (this implies that the mode selected is 1920x1080).
In short is there anyway to get the behavior if the combo libcamera and ffmpeg pipeline using just libcamera?
Statistics: Posted by wkeeling — Wed Nov 26, 2025 7:08 pm