Got another computer vision guide for you all, this time we are doing some pose estimation! We are using YOLO11 (fresh off the press) and a Pi 5 to draw keypoints and guess the position and pose of a person in the camera. Its really fun and we create a game of space invaders controlled by moving your head around! And with some optimisation we are able to get the FPS well into to 30+ range which is incredible for low powered hardware like the Pi!
Have you ever wanted to dive into computer vision? How about on a low-power and portable piece of hardware like a Raspberry Pi?
Well, in this guide we will be setting up some with OpenCV and the YOLO pose estimation model family on the Raspberry Pi…
Hello, I repeated the experiment according to this guide, and after running it, I found that the preview window did not appear, but there were running results, such as:[1:48:29.198092882] [8081] INFO Camera camera_manager.cpp:327 libcamera v0.4.0+53-29156679
[1:48:29.205834595] [8091] INFO RPI pisp.cpp:720 libpisp version v1.1.0 e7974a156008 27-01-2025 (21:50:51)
[1:48:29.215435934] [8091] INFO RPI pisp.cpp:1179 Registered camera /base/axi/pcie@120000/rp1/i2c@80000/imx708@1a to CFE device /dev/media1 and ISP device /dev/media2 using PiSP variant BCM2712_C0
[1:48:29.218445794] [8081] INFO Camera camera.cpp:1202 configuring streams: (0) 640x640-RGB888 (1) 1536x864-BGGR_PISP_COMP1
[1:48:29.218552390] [8091] INFO RPI pisp.cpp:1484 Sensor: /base/axi/pcie@120000/rp1/i2c@80000/imx708@1a - Selected sensor format: 1536x864-SBGGR10_1X10 - Selected CFE format: 1536x864-PC1B
0: 320x320 (no detections), 150.9ms
Speed: 3.6ms preprocess, 150.9ms inference, 0.7ms postprocess per image at shape (1, 3, 320, 320)
Do you know what is going on?
Are you able to enter the below command into the terminal on your Pi? It will indicate whether or not the camera itself is being detected by the Pi and display a 5 second feed from the camera if it is working.
