computer-vision

Recurrence Meets Transformers for Universal Multimodal Retrieval

🏠 OpenStreetMap, AI import tool for buildings in Poland

Summit Vitals: Multi-Camera and Multi-Signal Biosensing at High Altitudes

A web app using Vision Transformers for quick and accurate diagnosis of plant diseases, aiding effective treatment and plant health.

AI-powered disaster response platform with offline-first architecture using Gemma 3n. Provides computer vision hazard detection, voice analysis with emergency keywords, PDF report generation, and multi-user coordination - all working without internet access.

Engage in a semantic segmentation challenge for land cover description using multimodal remote sensing earth observation data, delving into real-world scenarios with a dataset comprising 70,000+ aerial imagery patches and 50,000 Sentinel-2 satellite acquisitions.

My implementation of the model KosmosG from "KOSMOS-G: Generating Images in Context with Multimodal Large Language Models"

building various transformer model architectures and its modules from scratch.

🦫 Open-source iOS AI companion with a live VRM avatar that sees you — bring-your-own-key chat, vision, speech-to-text & text-to-speech (OpenAI, Anthropic, Gemini, ElevenLabs, Groq & more). SwiftUI.

Unsupervised specificity-guided optimization of Image Captioning models to encourage meaningful diversity in the generated captions. Code for the paper Generating Diverse and Meaningful Captions: Unsupervised Specificity Optimization for Image Captioning (Lindh et al., 2018).

Deep Learning and Computer Vision application for chessboard detection and chess pieces classification.

Fast and practical image deraining with PyTorch.

基于Qwen2.5-VL-3B + QLoRA的智能驾驶场景结构化理解系统

A fine tuned yolov111 model to detect fire

Computer Vision project focused on detecting smoke and fire in wild environments. The Google Vision Transformer was fine-tuned on a custom dataset.

YOLOv26 and RF-DETR object detection for civilians, rescuers, and animals in disaster scenarios. Real-time support for emergency response in floods and structural collapses.

Caption images across your datasets with state of the art models from Hugging Face and Replicate!

ArtSeek: Deep artwork understanding via multimodal in-context reasoning and late interaction retrieval

[ICCV 2025] Know "No" Better: A Data-Driven Approach for Enhancing Negation Awareness in CLIP

The License Plate Recognition (LPR) System is a computer vision project designed to detect and recognize license plates in images or video streams.