Questions tagged [yolov4]

75 questions
1
vote
1 answer

Converting Yolov4 Tiny to tflite. error:cannot reshape array of size 372388 into shape (256,256,3,3)

i'm converting my custom weights file to tflite by using open source from https://github.com/haroonshakeel/tensorflow-yolov4-tflite. there is no error when i convert Yolov4.weights to tflite but when i switch to Yolov4-tiny.weights i got an error…
lin
  • 11
  • 1
1
vote
0 answers

YOLOv4 custom function doesnt detect video while running on GPU

im using YOLOv4 custom function from the AI guys's repo. Everything works flawlessly, including convert weights file to pb file, run detection on both image and videos, then this error shows up on both pc and colab: When i use only CPU, detection…
1
vote
3 answers

Text Detection with YOLO on Challenging Images

I have images that look as follows: My goal is to detect and recognize the number 31197394. I have already fine-tuned a deep neural network on text recognition. It can successfully identify the correct number, if it is provided it in the following…
spadel
  • 282
  • 2
  • 15
1
vote
0 answers

I get TypeError when using NMSBoxesRotated for Object Detection using YOLOV4

Error says: "TypeError: function takes exactly 3 arguments (4 given)" index=cv.dnn.NMSBoxesRotated(boxes_t,confidences,0.5,0.4) where boxes_t is tuple(x,y,w,h) values how to resolve this? Error image
SUMAN E N
  • 11
  • 1
1
vote
1 answer

YOLOv4 darknet weights

So I am training YOLOv4 using this command !darknet/darknet detector train darknet/data/obj.data darknet/cfg/yolov4-obj.cfg yolov4.conv.137 -dont_show -map and saving the weight files in my drive. Checking the drive I notice only best.weights and…
S-B
  • 11
  • 1
1
vote
1 answer

Failed to initialize detector. Input tensor has type kTfliteFloat32 ML Kit

I'd like to know if it is possible to implement a darknet model (yolov4-tiny) converted to .tflite in android studio with ML Kit. I've tried to use this repository: https://github.com/googlesamples/mlkit/tree/master/android/vision-quickstart , but…
1
vote
0 answers

Yolov4 weights conversion to tflite failure

I am using Google Colabs to convert Yolov4 darknet weights to tflite version. I used this blog to train my own yolo detector and I got to acceptable accuracy for my detections. Then I tried every repository on github 1,2,3 to covert my custom…
1
vote
0 answers

How to sort YOLOv4 bounding box?

I have trained the yolov4 from alexeyab darknet repo to detect characters in a number plate. It segments the character correctly but the bounding boxes are in random order. How can I sort the bounding box from top left to bottom right for image like…
0
votes
0 answers

Get the accuracy in scaled Yolov4

From this repo wang2020scaled YOLOv4-large branch, I used it for custom detecting bottles. I want to Generate the accuracy as this formula said My trying in test.py and utils/general.py which the function called ap_per_class responsiblle for…
memi
  • 15
  • 8
0
votes
0 answers

Yolov4 Custom Training - Couldn't open file: data/obj.data

The darknet training command darknet.exe detector train data/obj.data yolo-obj.cfg yolov4.conv.137 produces the error: Couldn't open file: data/obj.data on my Windows 10 device. I have followed the instructions in the AlexeyAB README file under "How…
AK_KA
  • 15
  • 5
0
votes
0 answers

Do you need to install Darknet on your machine to use YOLOv4 weights & .cfg in openCV

I'm trying to use OpenCV and YOLOv4 / YOLOv4 tiny weights and .cfg to make object detection predictions. My code doesn't want to run and keeps running into an error at modelConfiguration = "yolov4.cfg"; modelWeights = "yolov4.weights"; # Load…
0
votes
1 answer

YOLO: -bash: ./darknet: No such file or directory

When I try to train my yolov4 on my vm with command ./darknet detector train build/darknet/x64/data/obj.data cfg/yolov4-my_gray.cfg backup/yolov4-my_gray_1200.weights -gpus 0,1,2,3 -dont_show I received -bash: ./darknet: No such file or directory Of…
Mathew
  • 61
  • 5
0
votes
0 answers

Imutils.video FPS reads just the first frame of the video

I am working on a Yolo based object detector, and i want to deploy it on videos/ipcamera. i used the imutils.video module to loop through all the frames of the video. but the result shows just the detection on the first frame of the video. I want to…
0
votes
0 answers

Transfer leaning on custom YOLOv4 Model

I have an existing YOLOv4 weight file that is trained on my custom dataset. Can I use this model for transfer learning and use more dataset to improve it?If yes,it will be helpful if you brief the process.
0
votes
0 answers

Bad result on YoloV4 when multiple objects exist

I have an application requires to detect holes and cracks on surfaces. I followed yolov4 instructions to train my my own objects. The detection result is ok when it consist on only 1 objects. However, when the image consists of more than 1 detected…
JL2018
  • 53
  • 4
1
2 3 4 5