There is a huge component shortage from broken supply chains (especially during COVID) and challenging with e-recycling lead to massive waste and pollution. Our project automatically disassembles electronics to make reuse or recycling much easier, leading us to a more circular economy.
Prediction road sign by computer vision
AI-powered multimodal product price estimation using NLP, Computer Vision, and XGBoost.
Proxymow Server is a robotic lawn mower system that uses computer vision to locate and guide the mower.
Will utilise convolutional neural network (CNN) to develop a model to accurately detect brain tumours from MRI images.
This is a final project for CS 674: Intelligent Visual Computing.
Open Source Agricultural Embedding Evaluation API
An intelligent, multi-modal insurance claim automation engine for Health & Auto sectors. Features CNN-based vehicle damage assessment, Groq LLM adjudication, OCR document extraction, and a dual-interface (Web + Telegram).
This project trains a lightweight CNN on the CIFAR-10 dataset and goes beyond raw accuracy by focusing on interpretability. Using TensorFlow/Keras and Plotly, it provides interactive visualizations, per-class performance breakdowns, confidence histograms, and detailed grids of correct and wrong predictions to analyse how the model actually behaves.
The ReScape algorithm removes the perspective distortion from coral-reefscape images by transforming them into a top-down view.
End-to-end deep learning system for sugarcane leaf disease detection using PyTorch, ResNet-50, FastAPI, and Docker.
Classifier for the MNIST handwritten digit set.
Autonomous motion-tracking sentry for ESP32-CAM. Features real-time CV blob detection, WebSocket telemetry, and a custom futuristic Web HUD for kinetic manual override.
The project deals with augmentations, learning neuronal networks, Kaggle competition.
TasteRank: Personalized Image Search and Recommendation. This research project proposes an AI-based method for scoring photos on relevance to user interests. TasteRank leverages language and vision models, including Mistral LLMs and OpenAI’s CLIP, and applies multimodal machine-learning techniques.
NowFishing - 온디바이스 AI 기반 실시간 어종 판독 및 스마트 낚시 플랫폼
Drawing game judged by AI — draw the prompt, AI scores you
Coursework using CNN and Machine Learning - Image-Based Accident Severity Prediction for Safer Navigation in the UK: Enhancing Generative Routing Application such as Google Map or Waze using AI.
This project applies YOLOv8 for object detection, specifically for Automatic Number Plate Recognition (ANPR), with OCR powered by EasyOCR. It processes images and video files to detect license plates and recognize their text, making it useful for applications such as traffic analysis, smart parking, and surveillance.
Multimodal deep learning model for detecting fraudulent job postings using text and image fusion + Explainable AI Integration
To associate your repository with the computer-vison topic, visit your repo's landing page and select "manage topics."