Shakhizat Nurgaliyev
Embedded Engineer | Machine learning and Robotics enthusiast
- Report this post
Raspberry Pi 5 on Steroids! 🔥🔥🔥🔥The Raspberry Pi AI kit allows us to convert custom .onnx model to Hailo format. Below is an example of detecting a safety helmet using the Ultralytics YOLOv8n model with an impressive result, achieving the inference speed of 30FPS. You can follow tutorial to retrain you own custom model: https://lnkd.in/dF3jTixpHappy developing!Special thanks to Akin Victor, the talented creator of this video, which I used to demonstrate the capabilities of the Raspberry Pi AI kit.
183
15 Comments
Muhammad Rizwan Munawar
Computer Vision Engineer @Ultralytics | Solving Real-World Challenges🔎| Python | Published Research | Open Source Contributor | GitHub 🌟 | Daily Computer Vision LinkedIn Content 🚀 | Technical Writer VisionAI @Medium📝
10h
- Report this comment
Nice work Shakhizat! BTW, how many FPS RPI provides?
1Reaction 2Reactions
sulaiman mahmoud
Machine Learning Engineer and Data Scientist
2h
- Report this comment
Insightful!Are you tryinng to apllying model more than one class if yes how does it look and how it's behave , how many fps will be??
1Reaction
Kourosh Karimi
AI Engineer
6h
- Report this comment
that is great. what is the resolution of the input image and how much is the processing time (FPS)?
1Reaction
Sachin Chaudhary
MLE at Intelliroute | Programming language - C++, Java, python | Machine Learning | Deep Learning | YOLO | Certification- AI900, DP900 | SQL | HTML ,CSS, JS | J2EE | Docker
6h
- Report this comment
Very informative
1Reaction
Muhammad Fahad Qureshi
AI Engineer || ML Engineer || NLP || LLMs || Deep Learning || Data Scientist || Satellite Communication
9h
- Report this comment
Good to know!
1Reaction
Timothy Goebel
Transform Your Business with Cutting-Edge Computer Vision and Edge AI Solutions | AI Specialist & Innovator
14h
- Report this comment
What was resolution of the camera? Nice work.
3Reactions 4Reactions
Muhammad Awab Younas
CV intern@NRTC 💻 Programmer | 🤖 AI Enthusiast | 🧠 Computer Vision Practitioner | 🤖 Robotics Hobbyist | 🧠 Deep Learning Explorer | 📷 Camera Aficionado
8h
- Report this comment
Where to buy this hailo AI kit?
1Reaction
Muhammad Ibrahim
Affiliate marketing ||Digital Marketing ||social media marketing and ||Let's connect grow together
12h
- Report this comment
Very helpful!
1Reaction
Sidra Veli
Computer Vision Engineer /Medical image analysis
8h
- Report this comment
Very helpful!
1Reaction
To view or add a comment, sign in
More Relevant Posts
-
Shakhizat Nurgaliyev
Embedded Engineer | Machine learning and Robotics enthusiast
- Report this post
The story continues!Testing the Raspberry Pi AI Kit with a Hailo AI acceleration module. Here's an example video demonstrating instance segmentation with multiple objects using the Ultralytics YOLOv5 model, achieving an average inference speed of 30fps. CPU is not utilized for the inference.Thanks for the awesome video from London (c) 📷 Madiyar Aitbayev
382
16 Comments
Like CommentTo view or add a comment, sign in
-
Shakhizat Nurgaliyev
Embedded Engineer | Machine learning and Robotics enthusiast
- Report this post
I successfully ran Ultralytics' YOLOv6/8 object detection models on the Raspberry Pi 5 board with AI Kit powered by a Hailo AI acceleration module. The results are impressive, achieving an inference speed of 30fps, a performance that was not possible to achieve without the Hailo module before.Raspberry Pi Foundation, Ashley Whittaker, Seeed Studio, Jenna Zhu, Elaine Wu.
60
Like CommentTo view or add a comment, sign in
-
Shakhizat Nurgaliyev
Embedded Engineer | Machine learning and Robotics enthusiast
- Report this post
You might be wondering, is it possible to fine-tune large language models on small edge devices like the Nvidia Jetson Orin? Yes, it is possible. In just half an hour and for a cost of $10, you can generate your ownsynthetic dataset using OpenAI's ChatGPT. Then fine-tune LLM on the NVIDIA Jetson AGX Orin Developer Kit within just 2 hours ,and deploy the model using different tools like llamaspeak, NanoLLM and PiperTTS(Michael Hansen). A big thanks to Dustin Franklin and the NVIDIA Robotics Jetson team for providing these powerful tools!Here's a demo video of my work in action, an AI Assistant for Harry Potter Fans, which was fine-tuned on the Meta's Llama3-8B model. Read more on Hackster.io: https://shorturl.at/S4UDsJoin our Jetson AI Lab Research Group(https://lnkd.in/dQDu7vm6) Seeed Studio,NVIDIA Robotics, Johnny Núñez Cano, Elaine Wu, David Pinto, Mieszko Syty.
45
2 Comments
Like CommentTo view or add a comment, sign in
-
Shakhizat Nurgaliyev
Embedded Engineer | Machine learning and Robotics enthusiast
- Report this post
Well, managed to do it, Shawn Hymel 😅😀I also received a new dev board from OpenMV LLC. Hope to try deploying a model on it using Edge Impulse, David Tischler. Thanks again, Kwabena Agyeman 👍👍👍.
20
5 Comments
Like CommentTo view or add a comment, sign in
-
Shakhizat Nurgaliyev
Embedded Engineer | Machine learning and Robotics enthusiast
- Report this post
Wanna build a Raspberry Pi 5 cluster? @Ashley Whittaker
26
Like CommentTo view or add a comment, sign in
-
Shakhizat Nurgaliyev
Embedded Engineer | Machine learning and Robotics enthusiast
- Report this post
Happy #Pi Day! For Raspberry Pi enthusiasts))))I just upgraded my Beowulf Nvidia Jetson cluster by adding Distributed Ceph Storage cluster using the newly released Raspberry Pi 5 boards. Jeff Geerling Raspberry Pi Foundation Raspberry Pi NVIDIA Robotics #raspberrypi #ceph #cluster
55
2 Comments
Like CommentTo view or add a comment, sign in
2,982 followers
- 48 Posts
View Profile
FollowExplore topics
- Sales
- Marketing
- Business Administration
- HR Management
- Content Management
- Engineering
- Soft Skills
- See All