Skip to content
View PSR0001's full-sized avatar
🏠
Working from home
🏠
Working from home
7 followers · 7 following

Achievements

Achievement: Pull Sharkx2

Achievements

Achievement: Pull Sharkx2

Block or report PSR0001

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
PSR0001/README.md

👋 Hi, I'm Partha Singha Roy

Embedded Systems Engineer | Embedded Linux | Multimedia & Camera Systems | Edge AI

I build embedded software that connects hardware, multimedia pipelines, and intelligent edge applications.

My work focuses on Embedded Linux, camera systems, video processing, and hardware-accelerated AI, with hands-on experience developing applications on Rockchip RK3576 and Renesas RZ/V2L platforms.

What I Work On

  • Camera & Imaging: V4L2, camera pipelines, ISP, RKAIQ, 3A (AE/AWB/AF)
  • Embedded Multimedia: Rockchip MPP, H.264, DMA-BUF, zero-copy pipelines, RGA
  • Embedded Linux: Yocto, BitBake, U-Boot, Linux, device tree
  • Edge AI: YOLO, RKNN, NPU acceleration, on-device inference
  • Programming: C, C++, Python, JavaScript, CMake

Featured Projects

1. Dual-Camera Image Fusion

Rockchip RK3576 | C++ | OpenCV

Developing dual-camera image fusion using Gaussian-Laplacian pyramids, with a focus on embedded real-time image processing.

View Project

2. YOLO Hybrid Training on RK3576

C++ | RKNN | NPU | CPU

Exploring hybrid neural network training on resource-constrained embedded hardware, combining NPU matrix multiplication with CPU-based gradient computation and optimization.

View Project

3. Embedded AI & Multimedia Experiments

Working with camera capture, hardware-accelerated video processing, zero-copy buffers, and on-device object detection on Rockchip platforms.

Technology Stack

Languages

C C++ Python JavaScript

Embedded Linux

Yocto BitBake U-Boot Linux Device Tree

Camera & Multimedia

V4L2 Media Controller ISP RKAIQ Rockchip MPP RGA DMA-BUF

Edge AI

YOLOv5 YOLOv8 RKNN NPU Acceleration

Hardware Platforms

Rockchip RK3576 Renesas RZ/V2L

Currently Exploring

  • Linux Media Controller framework and camera pipeline architecture
  • ISP internals and camera 3A algorithms
  • Real-time embedded image processing
  • CPU + GPU + NPU heterogeneous computing
  • Native C++ AI applications on embedded Linux

Knowledge Sharing

I also run Embedded AI Lab, where I share practical experiments, demonstrations, and tutorials on embedded AI and hardware-accelerated computing.

YouTube: Embedded AI Lab

Let's Connect

I'm interested in opportunities involving:

  • Embedded Linux & Multimedia Software
  • Camera Systems & Imaging
  • Embedded AI & Edge Computing
  • SoC Software & Hardware Acceleration

Open to connecting with engineers, developers, and teams building next-generation embedded systems.

Pinned Loading

  1. msvsdspwm msvsdspwm Public

    VSD-Mixed-Signal-PD-Research

    Jupyter Notebook 1 4

  2. DSP-in-ESP32 DSP-in-ESP32 Public

    WebSocket Based Real Time DSP

    C

  3. EmbeddedAiLab/ai-camera EmbeddedAiLab/ai-camera Public

    JavaScript

  4. EmbeddedAiLab/dualcam_fusion EmbeddedAiLab/dualcam_fusion Public

    Contact Info embeddedailab.official@gmail.com

    Makefile

  5. EmbeddedAiLab/rag-retro-console-web EmbeddedAiLab/rag-retro-console-web Public

    C++

  6. EmbeddedAiLab/rk3576-yolo-hybrid-training EmbeddedAiLab/rk3576-yolo-hybrid-training Public

    Contact Info embeddedailab.official@gmail.com

    C++