darknet
yolov5
Our great sponsors
darknet | yolov5 | |
---|---|---|
62 | 128 | |
21,369 | 46,202 | |
- | 2.8% | |
7.0 | 8.9 | |
24 days ago | 5 days ago | |
C | Python | |
GNU General Public License v3.0 or later | GNU Affero General Public License v3.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
darknet
-
Anybody building ML models in C++?
YoloV3/4 is C based if that counts: https://github.com/AlexeyAB/darknet
-
[D] Fixing the angle of Skewed Paintings, see comments
This is all well-known information, see any (and all!) previous discussions when YOLOv5 comes up. For details: https://github.com/AlexeyAB/darknet/issues/5920
-
Viseron 2.0.0 - Self-hosted, local only NVR and AI Computer Vision software.
Yes it is official, it is in another repo however, but it is linked from the main repo https://github.com/AlexeyAB/darknet
- Machine learning Library in C?
-
GPL vs MIT.
Still to long. Here's my favourite license: https://github.com/AlexeyAB/darknet/blob/master/LICENSE
-
Pretrained model on aerial object detection
(As an aside, it isn't just that YOLOv5 is evil, the problem is YOLOv5 is both slower and less precise than YOLOv4, and even then, no-one can reproduce the performance results they claim to get.) Source: https://github.com/AlexeyAB/darknet/issues/5920
-
[D][P] YOLOv6: state-of-the-art object detection at 1242 FPS
For a quick start on YOLOv3 / 4 follow the instructions here: https://github.com/AlexeyAB/darknet
- Does Multi Object Tracking work better (precision/recall) on videos than jury rigging a SOTA image object detection to work on videos?
-
Which model is best for detecting small objects? Yolov3? MaskRCNN, Faster-RCNN?
YOLOv4-tiny-3L is what I use when I'm doing projects that require finding small objects. And for really tiny objects in large images where the object would be resized to zero (or near zero) due to network dimensions, then I use DarkHelp with tiling, combined with YOLOv4-tiny-3L. Depending on how precise the bounding boxes need to be, I also tend to use YOLOv4-tiny, which is faster but less precise than YOLO-v4-tiny-3L.
yolov5
-
Changing labels of default YOLOv5 model
I am using the default YOLOv5m6 model here with sahi/yolov5 library for my object detection project. I want to change just some of labels - for example when YOLO detects a human, I want it to label the human as "threat", not "person". Is there any way I can do it just changing some code, or I should train the model from scratch by just changing labels?
-
SS: Events triggered no matter the detection zone specified
What I did is hand off my detections to an Nvidia GPU running Yolov5 https://github.com/ultralytics/yolov5 which does the triggering if it's an object class I'm interested in. Since Synology is always caching at least 5 seconds of camera video the 50-100mS delay of grabbing a frame and analyzing before triggering recording is fine. Wish they'd implement something like this.
-
AMD ROCm 5.5 In The Process Of Being Released
this is about all I could find, maybe you've been getting different degrees of optimized python libs that do the image resizes. https://github.com/ultralytics/yolov5/issues/11469
-
good computer vision or deep learning projects in github
YOLOv5 (GitHub: https://github.com/ultralytics/yolov5) is a fast, accurate object detection model with code for training, testing, deployment, and pre-trained weights.
-
[D] Extracting the class labels and bounding boxes for objects, from a YOLO7 model after converting to an ONNX model
Those dimensions suggest you need to apply (i.e. roll your own) non-max suppresion to the outputs: relevant link
-
Thought Dump About Recent AI Advancements And Palantir
- YOLOv5 https://github.com/ultralytics/yolov5 (open source, so not Palantir's)
-
How to build computer vision dataset labeling team in-house
A team of annotators and the infrastructure described in this article I needed to label my dataset, which was collected from cameras on the road (30k+ photos). This dataset was necessary to train an object detection model on six classes: [person, car, bus, bicycle, motorcycle, truck]. I released the dataset, created in this manner, as open source, and it can be downloaded here (link) together with trained YOLOv5s and YOLOv5x models from a popular repository (link) using this dataset. The license is simple: "Use it well"!
-
Flutter Object Detection App + YOLOV5 Model.
Before you can use YOLOv5 in your Flutter application, you'll need to train the model on your specific dataset. You can use an existing dataset or create your own dataset to train the model. For this post I am using the pretrained model of yolov5 available on https://github.com/ultralytics/yolov5 as we are performing object detection we need to converts the pretrained model weights to torchscript format.
-
NotImplementedError (YOLOv5)
Thank you Link to the codefor taking the time to reply. I have modified the code as you suggested. And now I see the GPU being utilized. But the precision, recall, mAP is all zero. At least it displays as zero.
-
NVIDIA Jetson AGX Orin is now compatible with balena
Read more here about how Theia Scientific is currently using the newest NVIDIA Jetson AGX Orin on their fleet of microscopes. Theia Scientific and Volkov Labs, improved the Jetson AGX Orin inference speeds up to 30FPS. The Jetson AGX Orin running YOLOv5 tripled the Frames-per-Second (FPS) compared with the latest Jetson AGX Xavier.
What are some alternatives?
detectron2 - Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.
mmdetection - OpenMMLab Detection Toolbox and Benchmark
Deep-SORT-YOLOv4 - People detection and optional tracking with Tensorflow backend.
tensorflow-yolov4-tflite - YOLOv4, YOLOv4-tiny, YOLOv3, YOLOv3-tiny Implemented in Tensorflow 2.0, Android. Convert YOLO v4 .weights tensorflow, tensorrt and tflite
yolor - implementation of paper - You Only Learn One Representation: Unified Network for Multiple Tasks (https://arxiv.org/abs/2105.04206)
OpenCV - Open Source Computer Vision Library
yolov5-crowdhuman - Head and Person detection using yolov5. Detection from crowd.
CenterNet - Object detection, 3D detection, and pose estimation using center point detection:
yolov3 - YOLOv3 in PyTorch > ONNX > CoreML > TFLite
edge-tpu-tiny-yolo - Run Tiny YOLO-v3 on Google's Edge TPU USB Accelerator.
YOLOv6 - YOLOv6: a single-stage object detection framework dedicated to industrial applications.
gocv - Go package for computer vision using OpenCV 4 and beyond.