Code With Aarohi
Code With Aarohi
  • Видео 271
  • Просмотров 3 373 959
Florence-2 : Advancing a Unified Representation for a Variety of Vision Tasks | Paper Explained
Florence-2, a novel vision foundation model with a unified, prompt-based representation for a variety of computer vision and vision-language tasks.
Try out the Florence-2 model here: huggingface.co/spaces/gokaygokay/Florence-2
Paper: arxiv.org/pdf/2311.06242
Florence-2 is pre-trained on our FLD-5B dataset encompassing a total of 5.4B comprehensive annotations across 126M images.
#computervision #largelanguagemodels #languagemodels #microsoft #ai #artificialintelligence
Просмотров: 852

Видео

YOLOv8 on Jetson Nano Using DeepStream
Просмотров 69416 часов назад
DeepStream is developed by NVIDIA. DeepStream is a SDK that includes libraries, APIs, and pre-trained models for building and deploying AI-powered applications. GitHub: github.com/AarohiSingla/DeepStream-Yolov8-Jetson-Nano yolov4 weights: github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v3_optimal/yolov4.weights yolov4 cfg file: raw.githubusercontent.com/AlexeyAB/darknet/master/cfg/yo...
Steps to Recover a Hacked Gmail and YouTube Account
Просмотров 62521 час назад
Learn from my personal experience of recovering a hacked Gmail and RUclips account in this detailed video. I share step-by-step instructions on how I regained control of my accounts, including the exact procedures and actions I took. If you've been through a similar situation, this video will guide you through the recovery process effectively. Don't let account hacks disrupt your online presenc...
L-2 Create, manipulate and visualize tensors | Pytorch tensors
Просмотров 340День назад
In this video, you'll gain insights into the creation, manipulation, and visualization of tensors. GitHub: github.com/AarohiSingla/pytorch-for-deep-learning/tree/main/Lecture_2 For queries:- You can comment in comment section or you can email me at aarohisingla1987@gmail.com Tensors, fundamental to the field of machine learning and data science, are multidimensional arrays crucial for represent...
Automatic number plate recognition (ANPR) with Yolov9 and EasyOCR
Просмотров 515День назад
Automatic Number Plate Recognition (ANPR), also known as License Plate Recognition (LPR), is a technology that uses optical character recognition (OCR) to automatically read and interpret license plates on vehicles. For queries: You can comment in comment section or you can email me at aarohisingla1987@gmail.com ANPR technology is widely used for various purposes, including: Traffic Management:...
Client project - detect, track and classify Using YOLOv9 and ByteTrack
Просмотров 367День назад
About Project: Face Detection: Trained using YOLOv9. Custom Class Detection: Trained with YOLOv9, detecting 14 classes: ['firearm', 'knife/Sword', 'smoke', 'tank', 'warship', 'Missile', 'artillery', 'nudity', 'lingerie', 'aircraft', 'Cigarettes/Cigars/e-cigarettes', 'fire', 'cut/blood', 'bruise'] Classification: Frame-wise classification for violence, normal, and nudity classes, also using YOLO...
Image classification on Custom Dataset Using FasterViT
Просмотров 682День назад
Fast Vision Transformers with Hierarchical Attention Learn to perform Image classification with custom dataset using FasterViT model. GitHub: github.com/AarohiSingla/FasterViT Dataset added in GitHub repo: github.com/AarohiSingla/FasterViT/tree/main/RockPaperScissorsDataset Email: aarohisingla1987@gmail.com FasterViT FasterViT, a fast vision transformer model developed by NVIDIA. FasterViT (Fas...
FasterViT: Fast Vision Transformers with Hierarchical Attention
Просмотров 720День назад
In this tutorial, you will get a brief overview of FasterViT. Following that, I will demonstrate how to use a pre-trained FasterViT model on both images and videos. GitHub: github.com/AarohiSingla/FasterViT Email: aarohisingla1987@gmail.com #computervision #transformers #nvidia #imageclassification
NVIDIA NIM for Scaling Generative AI App Development
Просмотров 1 тыс.21 день назад
NVIDIA NIM for Scaling Generative AI App Development: 🚀 NVIDIA NIM @NVIDIAAIDev is the fastest way to deploy AI models on accelerated infrastructure across cloud, data center, and PC. 🔍 With just a few clicks, you can run models like MIXTRAL, GEMMA, and Llama. 🔗 Useful Links: • NVIDIA NIM Overview : nvda.ws/3yMBs7C • NVIDIA API Catalog : build.nvidia.com/explore/discover • Getting Started with ...
Train YOLOv10 on Custom Dataset
Просмотров 7 тыс.Месяц назад
Learn to perform custom object detection using YOLOv10. GitHub: github.com/AarohiSingla/YOLOv10-Custom-Object-Detection Dataset: Dataset is also present in GitHub repo. What is introduced in YOLOv10 : ruclips.net/video/2ZFJbeJXXDM/видео.html Email: aarohisingla1987@gmail.com YOLOv10: Real-Time End-to-End Object Detection Paper: arxiv.org/pdf/2405.14458 YOLOv10, developed by researchers at Tsing...
Learn What Is Introduced in YOLOv10 | YOLOv10 Paper Explained
Просмотров 6 тыс.Месяц назад
YOLOv10: Real-Time End-to-End Object Detection Paper: arxiv.org/pdf/2405.14458 YOLOv10, developed by researchers at Tsinghua University introduces a novel approach to real-time object detection. This version addresses deficiencies in both post-processing and model architecture found in earlier YOLO versions. By removing non-maximum suppression (NMS) and optimizing various model components, YOLO...
How to use Nvidia DeepStream with Jetson Nano | step by step tutorial
Просмотров 1,4 тыс.Месяц назад
DeepStream is developed by NVIDIA. DeepStream is a SDK that includes libraries, APIs, and pre-trained models for building and deploying AI-powered applications. Email id: aarohisingla1987@gmail.com Commands to execute : nvcc version This command will display the CUDA toolkit version installed on your system. dpkg -l | grep TensorRT This command will display the installed version of TensorRT on ...
Anomaly Detection: Explanation & Implementation
Просмотров 1,5 тыс.Месяц назад
This video will teach you - What is Anomaly detection? How Anomaly detection algorithm work? Implementation of Anomaly detection code of this paper: arxiv.org/pdf/1801.04264v3 GitHub: github.com/AarohiSingla/Anomaly-Detection Email id : aarohisingla1987@gmail.com The paper proposes a method for learning anomalies in surveillance videos without the need for annotating anomalous segments, which c...
Instance Segmentation Using YOLOv9 on custom dataset
Просмотров 2,1 тыс.Месяц назад
Learn how to perform Instance Segmentation Using YOLOv9 on custom dataset. GitHub: github.com/AarohiSingla/INstance-Segmentatio-Using-YOLOv9 Download custom dataset: universe.roboflow.com/asit-xno9q/levelup/dataset/4 Email id: aarohisingla1987@gmail.com Video on yolov9 architecture : ruclips.net/video/iH-c4_cjBbU/видео.html Instance Segmentation is a computer vision task, The goal of instance s...
How to Use NVIDIA ChatRTX - Create a Personal AI Chatbot on Your PC
Просмотров 753Месяц назад
What Is ChatRTX? ChatRTX is a demo app that lets you personalize a GPT large language model (LLM) connected to your own content-docs, notes, images, or other data. Leveraging retrieval-augmented generation (RAG), TensorRT-LLM, and RTX acceleration, you can query a custom chatbot to quickly get contextually relevant answers. And because it all runs locally on your Windows RTX PC or workstation, ...
Image Captioning using CNN and RNN | Image Captioning using deep learning
Просмотров 3 тыс.Месяц назад
Image Captioning using CNN and RNN | Image Captioning using deep learning
Object detection app using YOLOv8 and Android
Просмотров 12 тыс.2 месяца назад
Object detection app using YOLOv8 and Android
Object Detection Web Application with Flask and YOLOv9
Просмотров 5 тыс.2 месяца назад
Object Detection Web Application with Flask and YOLOv9
YOLOv9 on Jetson Nano
Просмотров 6 тыс.3 месяца назад
YOLOv9 on Jetson Nano
Track & Count Vehicles using YOLOv9 and ByteTrack
Просмотров 6 тыс.3 месяца назад
Track & Count Vehicles using YOLOv9 and ByteTrack
Object tracking using YOLOv9 and ByteTrack | Ultralytics
Просмотров 9 тыс.3 месяца назад
Object tracking using YOLOv9 and ByteTrack | Ultralytics
Detect and Track Objects using YOLOv9
Просмотров 1,1 тыс.3 месяца назад
Detect and Track Objects using YOLOv9
Automatic number plate recognition (ANPR) with Yolov9 and EasyOCR
Просмотров 10 тыс.3 месяца назад
Automatic number plate recognition (ANPR) with Yolov9 and EasyOCR
YOLOv9 vs YOLOv8 (Comparison on multiple videos)
Просмотров 5 тыс.3 месяца назад
YOLOv9 vs YOLOv8 (Comparison on multiple videos)
Vehicle and Pedestrian Detection Using YOLOv9 and Kitti dataset
Просмотров 1,4 тыс.3 месяца назад
Vehicle and Pedestrian Detection Using YOLOv9 and Kitti dataset
Oriented Bounding Boxes Object Detection | YOLOv8 OBB detection
Просмотров 3 тыс.3 месяца назад
Oriented Bounding Boxes Object Detection | YOLOv8 OBB detection
YOLOv9 Paper explained
Просмотров 7 тыс.3 месяца назад
YOLOv9 Paper explained
PPE Detection Using YOLO-World | Custom Object detection using YOLO-World
Просмотров 2 тыс.4 месяца назад
PPE Detection Using YOLO-World | Custom Object detection using YOLO-World
YOLOv9 on custom dataset | Object detection using YOLOv9
Просмотров 29 тыс.4 месяца назад
YOLOv9 on custom dataset | Object detection using YOLOv9
YOLO-World - Real-Time, Zero-Shot Object Detection
Просмотров 3,2 тыс.4 месяца назад
YOLO-World - Real-Time, Zero-Shot Object Detection

Комментарии

  • @usamarajput6418
    @usamarajput6418 9 часов назад

    this video helped me a lot. thank you Aarohi

  • @michaelzeuner1746
    @michaelzeuner1746 11 часов назад

    Why do you think the results from your test were so poor? Did you just need more training images?

    • @CodeWithAarohi
      @CodeWithAarohi 15 минут назад

      Yes, with more data, results will be better. Also, You can train for more epochs and try to use different learning rate and hyper parameters.

  • @poplu7076
    @poplu7076 18 часов назад

    while executing make -j4 i am getting error: desktop:/home/nvidia/opencv/build# make -j4 make; *** No targets specified and no makefile found. Stop. How to resolve this, PLeeeessee help 🙏🙏

  • @Bwajster
    @Bwajster 18 часов назад

    While I was trying to train the yolov8 model, I get an error: OSError: [Errno 12] Cannot Allocate Memory. Please help me resolve this. I'm training the yolov8 model on VS Code.

  • @RaghusuryaKonduru
    @RaghusuryaKonduru День назад

    while working in the server, not able to display the video. Can you make a video how can we do in the server.

  • @karthickkuduva9819
    @karthickkuduva9819 День назад

    Mam where can i find this room dataset

  • @user-tm7xd4im5m
    @user-tm7xd4im5m День назад

    Hi mam i need this code for practical implementation.... Nice video ❤

    • @CodeWithAarohi
      @CodeWithAarohi День назад

      Hello, This code is available for channel members (Contribution level 2)

  • @wieaswieas
    @wieaswieas День назад

    Please tell me how to connect a Q10F User Manual camera On Jetson nano because I tried and did not succeed, knowing that I used a video capture cable and it shows me /dev/video0 after executing ls /dev/video* knowing that my camera is video capture USB

  • @ankitaggarwal9195
    @ankitaggarwal9195 День назад

    Labels are empty in the dataset you have provided..?

  • @yabezD
    @yabezD День назад

    Kindly post a video for Deit

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    Hats off to your commendable efforts

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    Keep up this good work

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    👌👌

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    Very nicely explained

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    👏👏👏

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    Nice

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    Amazing

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    Very well explained

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    Awesome

  • @pifordtechnologiespvtltd5698
    @pifordtechnologiespvtltd5698 День назад

    Nice

  • @user-ff2tf3cw9j
    @user-ff2tf3cw9j День назад

    Hello Mam, i got this error how solve this error #NameError: name 'image_processor' is not defined

  • @rickyS-D76
    @rickyS-D76 День назад

    Thanks, do you have detailed video on video object detection with label and confidence score...or any other resource that can be helpful. Thank you.

  • @paulmobley9645
    @paulmobley9645 День назад

    Can you help me with a rural mountain community with problematic speeders on mountain roads?

  • @JohnSmith-gu9gl
    @JohnSmith-gu9gl День назад

    how did you come up with the values: [0.485, 0.456, 0.406] and [0.229, 0.224, 0.225] ?

  • @Disodimz
    @Disodimz 2 дня назад

    Whis is netter Yolov-9 or Florence-2

    • @CodeWithAarohi
      @CodeWithAarohi День назад

      Yolov9 is an object detection and segmentation model whereas Florence 2 is a vision language model. It can handle various tasks which yolov9 can't perform like Image captioning, text extracting etc.

  • @meghnadeshmukh4524
    @meghnadeshmukh4524 2 дня назад

    Thank you mam for teaching us sooo nicely.. I totally agree with @shahidulislamzahid... mam you are too good.

    • @CodeWithAarohi
      @CodeWithAarohi 2 дня назад

      Glad my videos are helping you. keep learning :)

  • @digambar6191
    @digambar6191 2 дня назад

    Thank you mam

  • @hxxzxtf
    @hxxzxtf 2 дня назад

    🎯 Key points for quick navigation: 00:05 *📚 Florence-2 is a lightweight vision language model that can handle various tasks based on simple instructions.* 00:26 *💡 The key innovation of Florence-2 is its ability to handle tasks like object detection, captioning, and detailed image analysis using a unified approach.* 04:13 *🔍 In computer vision, models need to understand both global concepts and finer details to be effective across different tasks.* 04:54 *📍 Spatial hierarchy refers to the understanding of visual information at different scales or levels of detail within an image.* 06:03 *🔎 Semantic granularity refers to how much detail we can understand from visual information, ranging from general ideas to specific details.* 09:11 *🤝 Multitask learning involves teaching a model to do multiple related tasks at the same time to improve its overall understanding and performance.* 10:08 *💪 Universal representation learning means training a single model that can understand different types of information without processing has several phases for ensuring correct and complete annotations.* 20:39 *👀 The detailed annotation process ensures that the FLD 5B data set is properly labeled across different levels of granularity, enhancing its utility for advanced AI applications.* Made with HARPA AI

  • @ahmadroyyan8523
    @ahmadroyyan8523 2 дня назад

    python version?

  • @ashutoshshukla4680
    @ashutoshshukla4680 2 дня назад

    Can we install i Jetpack sdk 5 in jetson nano

    • @CodeWithAarohi
      @CodeWithAarohi 2 дня назад

      No

    • @ashutoshshukla4680
      @ashutoshshukla4680 2 дня назад

      @@CodeWithAarohi were you able to use yolov8 in jetson nano

    • @CodeWithAarohi
      @CodeWithAarohi 2 дня назад

      Yes but for that I upgraded the python version to 3.8 and pytorch was compiled with cpu. Recently, I did another video where I was able to use yolov8 on jetson nano using deepstream.

    • @ashutoshshukla4680
      @ashutoshshukla4680 2 дня назад

      @@CodeWithAarohi i tried getting gpu inference for yolov8 but failed curious if you were able to, thanks for replying 🙏🙏

    • @CodeWithAarohi
      @CodeWithAarohi 2 дня назад

      @@ashutoshshukla4680 with deepstream, I was able to use jetson nano gpu

  • @ykakde
    @ykakde 3 дня назад

    Nicely explained mam😍😍😍😍

  • @varunsharma1889
    @varunsharma1889 3 дня назад

    But how did the hackers manage to change your GMail password even though you have 2 factor authentication? Did they get your browser session cookies somehow ?

  • @emirhanbilgic2475
    @emirhanbilgic2475 3 дня назад

    I was waiting for this video! Thank you!

  • @soravsingla8782
    @soravsingla8782 3 дня назад

    Nice

  • @balveersingh3051
    @balveersingh3051 3 дня назад

    Thankyou

  • @sahil5124
    @sahil5124 3 дня назад

    great explanation!

  • @user-ep2ci4mu8j
    @user-ep2ci4mu8j 3 дня назад

    AI renewed so soon!

    • @CodeWithAarohi
      @CodeWithAarohi 3 дня назад

      Absolutely! Technology moves fast these days.

  • @mohammadyahya78
    @mohammadyahya78 3 дня назад

    amazing explanation as usual

  • @mohtadianaqvi454
    @mohtadianaqvi454 3 дня назад

    Hi, Aarohi your videos are really helpful. I am facing the following error whenever I use model.fit function. model.fit(train_x, train_y, epochs = 6, batch_size = 10) Could you plz help why I am having this error and how to remove it? Epoch 1/6 --------------------------------------------------------------------------- ValueError Traceback (most recent call last) <ipython-input-98-4a0e0019ca96> in <cell line: 1>() ----> 1 model.fit(train_x, train_y, epochs = 6, batch_size = 10) 1 frames /usr/local/lib/python3.10/dist-packages/keras/src/backend.py in categorical_crossentropy(target, output, from_logits, axis) 5569 5570 """ -> 5571 target = tf.convert_to_tensor(target) 5572 output = tf.convert_to_tensor(output) 5573 target.shape.assert_is_compatible_with(output.shape) ValueError: Attempt to convert a value (SparseTensor(indices=tf.Tensor( [[0 0] [1 1] [2 1] [3 2] [4 1] [5 1] [6 2] [7 0] [8 1] [9 2]], shape=(10, 2), dtype=int64), values=tf.Tensor([1. 1. 1. 1. 1. 1. 1. 1. 1. 1.], shape=(10,), dtype=float32), dense_shape=tf.Tensor([10 3], shape=(2,), dtype=int64))) with an unsupported type (<class 'tensorflow.python.framework.sparse_tensor.SparseTensor'>) to a Tensor.

  • @karthickkuduva9819
    @karthickkuduva9819 3 дня назад

    Mam. Freezing all layers except the final classification layer is called transfer learning. And customized with our own dataset so it's also fine tuned model. The way I understand is correct ?

  • @shivkr21
    @shivkr21 3 дня назад

    hello ma'am , I couldn't find the c3d.pickle file , kindly help

    • @CodeWithAarohi
      @CodeWithAarohi 3 дня назад

      drive.google.com/drive/folders/1rhOuAdUqyJU4hXIhToUnh5XVvYjQiN50?usp=sharing

  • @user-ff2tf3cw9j
    @user-ff2tf3cw9j 3 дня назад

    Hello Mam can you make a video related to the : ## GCN-FFNN: A two-stream deep model for learning solution to partial differential equations

  • @luqmanabidoye7344
    @luqmanabidoye7344 3 дня назад

    The github code does not match the one shown in the video. No: step 2.... line in the code

  • @abubakarsaleem5167
    @abubakarsaleem5167 3 дня назад

    Ma'am, you have edited the yolov9.yaml file in models/detect/yolov9.yaml, but the pretrained model you are using is different, specifically "yolov9-e.pt." I may be wrong, but I think it is the "yolov9-e.yaml" file that should be edited instead. Btw, thanks for this informative tutorial.

  • @ykakde
    @ykakde 3 дня назад

    😍😍😍😍😍😍😍😍 very well explained mam

  • @ykakde
    @ykakde 3 дня назад

    😍😍😍😍

  • @abubakarsaleem5167
    @abubakarsaleem5167 3 дня назад

    So much helpful. Waiting for the second part

  • @mohammadyahya78
    @mohammadyahya78 4 дня назад

    I have a metal storage tank that is part of a water heater. During production of tank, leaking happen so early detection using either image processing or computer vision would be great to drop that storage tank and never let it through. I tried thermal image processing and contrast lighting but it's very difficult to see the leaking due to the water spraying and the coating around it that makes it difficult to find potential leaking holes. The video shows employee spraying water onto a pressured storage tank. Even when employee detect leaking while spraying, nothing can be detected from camera still. Any help with that please? Not sure if this one or another could work?

  • @likeyo-yy3vj
    @likeyo-yy3vj 4 дня назад

    great video,but why the fps so slow?

    • @CodeWithAarohi
      @CodeWithAarohi 3 дня назад

      I am also recording the screen from jetson nano to share the output. Other process are running... that's why

  • @businessgamerprb5398
    @businessgamerprb5398 4 дня назад

    I always have problem with OCR... I tried 3 different methods... Every time the region of number plate will be properly detected but then when I apply OCR on the cropped images... It just extracts #&((":! +3'srhed'hjhdv₹₹+:"+ This has given me better results than others but still OCR is detecting only about 50 percent characters correct... Idk what might be the issue