Autonomous spec-driven development plugin for Claude Code. Describe a feature, get merged, tested, reviewed code. Full pipeline from PRD to PR with dependency-aware task tracking, AI code review, and zero dependencies beyond bash + jq.

This is my Object Detection project using YOLOv8 on a COCO subset. I prepared the dataset, trained the model, performed inference and visualization, and evaluated its performance. The repo demonstrates the complete workflow with results and metrics for effective object detection.

Search Cocoapods in Alfred 2

No description available

Creating a CLIP training workflow based on the ViT-B Architecture (Framework: Open CLIP). Dataset used: laion-coco-nllb.

No description available

my own cocoapods-workflow

No description available

Alfred workflow to search CocoaPods from Alfred. Opens the library URL and copies the dependency string to the clipboard so that you can paste it into the Podfile directly.

No description available

Slice large COCO-annotated images into smaller tiles with adjusted annotations for training small-object detection models.

Dataset tools and annotation workflow for computer vision. Uses fiftyone, coco-annotator and CVAT.

No description available

End-to-end RF-DETR segmentation workflow: YOLO-to-COCO conversion, 50-epoch training, and batch prediction export.

Bidirectional YOLO-COCO annotation converter with integrated SAHI image slicing for object detection datasets. Jupyter notebook with step-by-step workflow.

Designed a scalable labeling workflow supporting 75+ object classes, annotation editing, and COCO-style JSON exports for defense systems and civilian protection.

WHR OAM + COCO Y0 Dashboard is a web-based analytical tool that reproduces the Excel-based World Happiness Report OAM workflow, automates COCO Y0 estimation, computes ranking deltas and correlations, and visualizes results through interactive tables and world maps.

Training LLMs and vision models. Supports COCO-format annotations, YOLOv8 training, and custom dataset workflows using annotation tools. Ideal for fine-tuning and building custom models with NLP or vision-language tasks

Automated pipeline for creating custom datasets from 3D assets using Blender. Includes scripts for rendering, annotation, and export to formats suitable for deep learning workflows (e.g., YOLOv5, COCO). Designed for reproducibility, scalability, and beginner-friendly customization.

A custom 50-image dataset annotated with CVAT and exported in YOLO 1.1 and COCO 1.0 formats, featuring five diverse classes: Person, Food, Vehicle, Animal, and Electronic Device. Created to demonstrate high-quality data labeling workflows for computer vision applications. ✅

This repository covers the complete workflow of object detection using Detectron2. It includes theoretical understanding, dataset annotation and COCO conversion, model training and evaluation, inference on new images, and an end-to-end wildlife monitoring pipeline addressing real-world challenges like occlusion and low-light detection.

A computer vision system that performs real-time face recognition and object detection using OpenCV, COCO dataset and Python. Includes customer identification, balance deduction logic, and automated receipt generation for a retail automation workflow. Developed as part of my graduation project with performance improvements, database integration.

PyTorch image captioning project built on the MS COCO dataset using a pretrained ResNet-18 encoder and LSTM decoder. This documents the end-to-end workflow for dataset preparation, model training, inference, and optional validation-based caption evaluation, with notebook outputs showing qualitative predictions and BLEU, ROUGE-L, and CIDEr results.