- Видео 5
- Просмотров 70 764
good rest
Добавлен 6 ноя 2017
Vision Transformers for Dense Prediction (ICCV 2021)
Vision Transformers for Dense Prediction (ICCV 2021) - State f the art Real-time (30 FPS and higher) neural network for Semantic segmentation and Mono-Depth estimation from one RGB image. DPT (Dense Prediction Transformers): github.com/intel-isl/DPT
René Ranftl, Alexey Bochkovskiy, Vladlen Koltun
openaccess.thecvf.com/content/ICCV2021/html/Ranftl_Vision_Transformers_for_Dense_Prediction_ICCV_2021_paper.html
René Ranftl, Alexey Bochkovskiy, Vladlen Koltun
openaccess.thecvf.com/content/ICCV2021/html/Ranftl_Vision_Transformers_for_Dense_Prediction_ICCV_2021_paper.html
Просмотров: 5 525
Видео
DPT: State of the art neural network for Segmentation and Depth estimation
Просмотров 1,8 тыс.3 года назад
DPT (Dense Prediction Transformers): State of the art real-time neural network for Semantic segmentation and Mono-Depth estimation from a single RGB image. paper: arxiv.org/abs/2103.13413 source code: github.com/intel-isl/DPT
Scaled YOLO v4 - The best neural network for object detection (CVPR 2021)
Просмотров 16 тыс.3 года назад
Scaled YOLO v4 - The best neural network for object detection: * Pytorch: github.com/WongKinYiu/ScaledYOLOv4 * Darknet: github.com/AlexeyAB/darknet * Paper: arxiv.org/abs/2011.08036 * Medium: alexeyab84.medium.com/scaled-yolo-v4-is-the-best-neural-network-for-object-detection-on-ms-coco-dataset-39dfa22fa982?source=friends_link&sk=c8553bfed861b1a7932f739d26f487c8 YOLOv4
YOLOv4 - The most accurate real-time neural network for object detection
Просмотров 46 тыс.4 года назад
YOLO v4 source code: github.com/AlexeyAB/darknet paper: arxiv.org/abs/2004.10934 medium: medium.com/@alexeyab84/yolov4-the-most-accurate-real-time-neural-network-on-ms-coco-dataset-73adfd3602fe?source=friends_link&sk=6039748846bbcf1d960c3061542591d7
YOLOv4 608x608 - New York - The most accurate real-time object detector
Просмотров 1 тыс.4 года назад
source code: github.com/AlexeyAB/darknet paper: arxiv.org/abs/2004.10934
Great talk. Thanks for sharing :)
Great presentation
Great talk
Hello, I just have a question Is GPU Nvidia gt 840m compatible for yolov4 ?????
Why not
We are building the cage that we will all live in soon. Nightmarish things are coming. Your inventions will be used against you.
Очень годно
Big Brother is born !
Excellent work!
bravo woosal
Still there is too much false positive errors, but getting better of course :)
Because this is a test in the most difficult conditions, rather than the simplest videos that are used to demonstrate other detectors. Tesla Autopilot detector will go crazy in this video, moreover, Tesla stumbles even in this case: twitter.com/FSD_in_6m/status/1400207129479352323
Hi , how can I run scaled yolov4 on jetson nano ?
Can u please suggest some good gpu?
for training i recommend a tesla m60 or if youre on budget tesla k80
Cyberpunk 2077
The Tensorflow version of YOLO does not support v4 ...
TensorFlow supports YOLOv4: github.com/hunglc007/tensorflow-yolov4-tflite and github.com/tensorflow/models/tree/master/official/vision/beta/projects/yolo But TensorFlow doesn't support ScaledYOLOv4 yet: github.com/hunglc007/tensorflow-yolov4-tflite/issues/280
music?
Monolith (Arnold mix) Pyramid
Which GPU run code??
What is the differences?? The accuracy?
alexeyab84.medium.com/scaled-yolo-v4-is-the-best-neural-network-for-object-detection-on-ms-coco-dataset-39dfa22fa982?source=friends_link&sk=c8553bfed861b1a7932f739d26f487c8
Scaled YOLOv4 :) Thanks for this great work.
congratulations!
Очень годно
Amazing! Would I use this video in the lecture material with reference?
no
awesome
your detection is so first.. bt i have very slow ... is that my computer issue ??
You must run it on CPU. You should use GPU to get faster results!
@@tafikelmehdi2913 can u send to me the code source of yolov4
@@noussaa1638 You can check this tutorial ! it was helpful for me ruclips.net/video/10joRJt39Ns/видео.html&ab_channel=TheAIGuy
@@tafikelmehdi2913 thanks
Compare to yolov3, which one faster?
Yolov4 is faster and more accuracte than Yolov3: user-images.githubusercontent.com/4096485/84686850-62001c00-af45-11ea-80b8-6d296bcc0a64.png Just set lower width and height in yolov4.cfg file: Yolov4 width=416 height=416, speed = 96 FPS, accuracy = 41% AP Yolov4 width=512 height=512, speed = 83 FPS, accuracy = 43% AP Yolov3 width=608 height=608, speed = 75 FPS, accuracy = 33% AP user-images.githubusercontent.com/4096485/82736855-43877600-9d35-11ea-972b-d09a055ce65f.png user-images.githubusercontent.com/4096485/82763581-874fad80-9e11-11ea-8719-82e44183658c.png
Do you have the code available for this?