Skip to content
View dronefreak's full-sized avatar
🎯
Focusing
🎯
Focusing
81 followers · 9 following

Achievements

Achievement: Pair ExtraordinaireAchievement: QuickdrawAchievement: Pull Sharkx2Achievement: YOLOAchievement: Starstruckx2Achievement: Arctic Code Vault Contributor

Achievements

Achievement: Pair ExtraordinaireAchievement: QuickdrawAchievement: Pull Sharkx2Achievement: YOLOAchievement: Starstruckx2Achievement: Arctic Code Vault Contributor

Block or report dronefreak

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. Markdown is supported. 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
dronefreak/README.md

Hare Krishna 🙏

I'm Saumya Kumaar Saksena

Perception Architect · Quantizing models for the edge · Shipping perception to production · ISO 26262 / SOTIF for autonomy

🚀  About Me

  • 5+ years building computer vision and deep learning systems across autonomous mobility, healthcare, and AR/VR — currently in Paris on safety-critical perception for self-driving shuttles
  • Hands-on with edge optimization and model quantization for low-power embedded devices, from training to on-device validation

🛠  Tech Stack

Category Stack
Languages & Tooling Python C++ MATLAB Simulink Bash YAML Pytest
Deep Learning & Perception PyTorch TensorFlow Keras OpenCV NumPy Matplotlib MLflow
Edge, Robotics & Deployment ROS / ROS 2 RTMaps CUDA TensorRT ONNX Docker Linux
DevOps & Collaboration GitLab Bitbucket GitHub Jenkins Jira Confluence Slack

⚙️  GitHub Analytics

Contribution snake animation

🤝🏻  Social Profiles

LinkedIn     Gmail     Google Scholar     Portfolio     Hugging Face

Pinned Loading

  1. human-action-classification human-action-classification Public

    Human action classification system with pose-based (MediaPipe) and video-based (3D CNN) models. Features 100+ architectures for real-time pose classification and temporal models pretrained on UCF-1…

    Python 269 68

  2. VisDrone-dataset-python-toolkit VisDrone-dataset-python-toolkit Public

    VisDrone aerial object detection toolkit with 33 models (Torchvision + YOLO), training, evaluation, video inference, benchmarking, and annotation conversion.

    Python 82 32

  3. dji-tello-object-detection-segmentation dji-tello-object-detection-segmentation Public

    Modern instance segmentation for DJI Tello drones with YOLOv8/Detectron2. Real-time performance, modular design, autonomous tracking.

    Python 69 21

  4. dji-tello-collision-avoidance-pydnet dji-tello-collision-avoidance-pydnet Public

    Real-time monocular depth estimation for DJI Tello drone collision avoidance. Modern TF2.x implementation with webcam support, comprehensive tests, and autonomous navigation. PyDNet-based with MiDa…

    Python 55 17

  5. dji-tello-target-tracking dji-tello-target-tracking Public

    Modern autonomous drone tracking using YOLOv8 deep learning and PID control for DJI Tello drones with real-time object detection and multi-target tracking.

    Python 55 17

  6. CABiNet CABiNet Public

    CABiNet: Efficient Context Aggregation Network for Low-Latency Semantic Segmentation (ICRA2021)

    Python 27 10