All Questions
11 questions
0
votes
0
answers
49
views
How to determine the orientation of a marker with tensorflow lite
I want to get the orientation (in 3d space if possible) of a marker object in a video stream using tensorflow lite on a coral tpu. My marker looks like:
The marker will be sticked on various objects ...
0
votes
0
answers
85
views
How do I install opencv on my Raspberry Pi 4 properly?
I kept getting this error.
ImportError: /usr/local/lib/python3.9/dist-packages/cv2/cv2.abi3.so: undefined symbol: __atomic_store_8
I've checked the python and opencv installed on my raspberry pi and ...
- 1
0
votes
0
answers
151
views
Speed Detection using TensorFlow Lite in a Raspberry pi 4B
i'm trying to create a vehicle collision detection system for my thesis project but i need to detect the speed of objects and distance to calculate the possible collisions. i'm trying to create it ...
- 1
-1
votes
1
answer
2k
views
I am trying to use mediapipe for hands detection and it is not working. Plus I am new to Python [closed]
INFO : Created TensorFlow Lite XNNPACK delegate for CPU
It is not detecting the hands or doing anything, only the camera is opening.
Code :
import cv2
import mediapipe as mp
cap = cv2.VideoCapture(0)
...
- 13
0
votes
0
answers
86
views
How can I extract a specific data from a function?
I printed a variable print(detection_result) which I called from detection_result = detector.detect(input_tensor) and it printed
DetectionResult(detections=[Detection(bounding_box=BoundingBox(origin_x=...
0
votes
1
answer
1k
views
MoveNet Pose Estimation renders inaccurate keypoints
I'm trying to run the MoveNet Pose Estimation model on a video but for some reason my keypoints are very inaccurate. I assume this does not have anything to do with the predictions itself but with how ...
- 7,914
0
votes
1
answer
1k
views
How to Resize the Tensorflow Output Bounding Box to Original Image
So, I am working on a Model, that I have successfully trained, it is a .tflite model and uses Tensorflow Lite. I am using Python 3 as my interface with tensorflow and I am unable to add a bounding box ...
-1
votes
1
answer
2k
views
Is there an alternative to the cv2.resize() function that is faster at upscaling an image?
Problem
I want to display an image fullscreen (1920 x 1080) at 60 FPS in my application, but what I'm actually getting is closer to 15 FPS.
What I've tried
This is the most simplified version of my ...
24
votes
4
answers
115k
views
(-5:Bad argument) in function 'rectangle' - Can't parse 'pt1'. Sequence item with index 0 has a wrong type
As I detect my tflite file, the problem happened.
The command I wrote.
python detect.py --weights ./checkpoints/yolov4-tiny-tf.tflite --size 416 --model yolov4 --image D:\yolov4\training\tensorflow-...
- 395
0
votes
0
answers
262
views
Backlight recognition of car using tflite and open cv
I am trying to detect the traffic light and backlight of a car using tensor flow lite for implementation on raspberry pi 3B module. I am using coco_ssd_mobilenet sample model for detection.
I have ...
1
vote
1
answer
595
views
Crop rectangle & show image that cropped in OpenCV Python
I'm new on opencv, I have an original image like this
and I want to crop each licence plate that i've cropped it with this code.
import re
import cv2
import numpy as np
# import tensorflow.lite as ...
- 145