All projects (64)
ThunderGP enables data scientists to enjoy the performance of FPGA-based graph processing without compromising programmability.
Exploit Xilinx FPGA's hardware to train neuroevolved binary neural networks, then solve Reinforcement Learning problems (like Atari Games).
We allow combining the power of FPGA acceleration with the ease of Python programming to use 100s of FPGAs for big data analytics.
A computer vision project focus on robust reconstructing the object digital model with real-time by simultaneous multi-view camera.
Fruit detection at the edge using a ZCU104.
Using the Ultra96-V2 board to detect person and make dormitory safer
This Project will make use of Xilinx Software to help educate visually impaired students to convert text data into audio
Designing an autoencoder based compression technique to compress images at near realtime 30 fps.
We are here going to build a power-optimized FPGA Hardware-based Agriculture Robotic Platform for Detecting and Removing Weed from the field
This project tries to achieve mapping of brain activity to some useful action.
Use an Ultra96V2 Board, a Yolov3 neuronal network running on a Xilinx DPU to calculate the perfect pizza cutting angle based on the toppings
End-to-end demonstration of 3D object detection in LiDAR point clouds using a deep neural network running on the ULTRA96V2.
Vision-based System designed to improve the shopping experience and to monitor social distancing.
Current ADAS algorithms such as object and lane detection perform badly in night conditions. How about enhancing the image as preprocessing.
A scalable solution for distributed Hot Desking Office management using the Ultra96-V2 for AI and image processing on the edge.
Join me in my journey to build an intelligent device for Wildlife biologists/conservationists using Xilinx Ultrascale FPGAs.
Detecting Covid-19 from X-Ray images using CNNs on cloud FPGAs
We implemented BERT large model inference Vitis HLS kernel on Alveo U50 card with Apach TVM S/W.
Quadcopter control and pole balancing using Deep Reinforcement Learning and Hand Gestures on Ultra96
Telehealth platform for remote diagnostics and treatment. Patient deployed or satellite center deployed tools. Open system tools.
System to fill the gunfire data gap and detects, locates gunfire to enable a fast and precise response to over 90% of gunfire incident.
In this work we design a generic FPGA accelerator design aiming at accelerating BNN neural networks with high performance in accuracy.
Creating spacially aware AI using a RealSense Camera and the Ultra96v2.
Deploy an object detection model on DPU to build a system which can show detected commodities in VCU decoded video or images from camera.