fcakyon

View JSON Representation

helping AI's to understand videos at @ultralytics & @viddexa

Funding Links: https://github.com/sponsors/fcakyon

• Name: fatih akyon
• Location: Ankara, Turkiye
• Company: @viddexa @ultralytics
• Kind: user
• Followers: 599
• Following: 65
• Total stars: 2445
• Repositories count: 131
• Created at: 2022-11-15T05:53:37.071Z
• Updated at: 2025-03-28T13:05:46.270Z
• Last synced at: 2025-03-28T13:05:46.270Z

GitHub Sponsors Profile

Maintaining my open-source packages takes a huge amount of time.
More sponsors = More motivation 🚀
🧐 about me

Passionate about Machine Learning/Deep Learning, Computer Vision
Ph.D. cand. in Informatics (multi-modal video understanding), BSc and Msc. in Engineering (deep learning/computer vision)
Loving research (📈 200+ citations) and open source (📈 2 million+ downloads)

🔥 some of my opensource works:
A lightweight vision library for performing large scale object detection/ instance segmentation:

https://github.com/obss/sahi

Packaged version of ultralytics/yolov5 + many extra utilities:

https://github.com/fcakyon/yolov5-pip

Deep learning based content moderation from text, audio, video & image input modalities:

https://github.com/fcakyon/content-moderation-deep-learning

Easiest way of fine-tuning HuggingFace video classification models:

https://github.com/fcakyon/video-transformers

Easy-to-use, class-balanced, cross-entropy and focal loss implementation for Pytorch:

https://github.com/fcakyon/balanced-loss

A simple Streamlit Component to compare images with a slider in Streamlit apps:

https://github.com/fcakyon/streamlit-image-comparison

Packaged, Pytorch-based, easy to use, cross-platform version of the CRAFT text detector:

https://github.com/fcakyon/craft-text-detector

Make your own dataset for object detection and transform the format to coco json format:

https://github.com/fcakyon/labelme2coco

📜 some of my published/submitted papers:

Deep Architectures for Content Moderation and Movie Content Rating
(Preprint)

Sequence Models for Drone vs Bird Classification
(Preprint)

Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection
(Accepted at ICIP 2022)

Drone-vs-Bird Detection Challenge at IEEE AVSS2021
(Published at AVSS 2021)

Track Boosting and Synthetic Data Aided Drone Detection
(Published at AVSS 2021)

Automated question generation and question answering from Turkish texts
(Accepted at SCI Journal, 2022)

Instagram Fake and Automated Account Detection
(Published at INISTA 2019)

• Current Sponsors: 2
• Past Sponsors: 0
• Total Sponsors: 2
• Minimum Sponsorship: $1.00

Featured Works

Active Sponsors

Past Sponsors

Sponsor Breakdown

• User: 2