search

agherzan / meta-raspberrypi

Yocto/OE BSP layer for the Raspberry Pi boards
https://www.yoctoproject.org/
MIT License
507 stars 397 forks source link

Install object_detect_tf post processing stage #1214

Open vivien opened 11 months ago

vivien commented 11 months ago

With this PR, one can add the meta-neural-network layer to their system, enable tensorflow support in libcamera-apps with e.g.:

PACKAGECONFIG:pn-libcamera-apps:append = " tflite"

and try object detection with the new libcamera-detect app or even libcamera-still --post-process-file /usr/share/libcamera-apps/assets/object_detect_tf.json --lores-width 400 --lores-height 300 for example.

(tested on raspberrypi4-64 with a Camera Module 3)

vivien commented 10 months ago

tensorflow-lite from meta-neural-network currently only builds for 64-bit systems. Please let me know what I must add in a new commit to skip the build of tensorflow-lite for 32-bit systems (libcamera-apps will build just fine if tensorflow was requested by the user but not available in sysroot.)

vivien commented 9 months ago

I've documented the upstream issue in the code and commit message and skipped the build on 32-bit systems (libcamera-apps builds fine when tflite isn't available in sysroot.)