Kukil
Processing long documents with VLMs or LLMs poses a fundamental challenge: input size exceeds context limits. Even with GPUs, as large as 12 GB can barely process 3-4 pages at
DeepSeek AI just unleashed DeepSeek OCR, a 3B-param beast that compresses entire documents into 100 vision tokens and still nails 97% accuracy.
Deploying ML on Arduino Nano 33 BLE. Explore TinyML techniques, setup steps, and why older Arduinos still rival the new Arduino Uno Q.
Learn how to build AI agent from scratch using Moondream3 and Gemini. It is a generic task based agent free from application APIs.
Learn how to setup a pipeline to run VLM on Jetson Nano using Huggingface Transformers. Run models like LiquidAI, Moondream2, FastVLM, and SmolVLM.
Testing Vision Language Models (VLM) on edge devices. Check how small VLMs perform on our custom Raspberry Pi Cluster and Jetson Nanos.
Explore the world of drone programming with computer vision! Maximize drone performance and precision for advanced applications using Python.
Learn how to create a pip installable package. Pip, PyPi, and various build backends. We will use PyOpenAnnotate as a reference for python packaging.
PyOpenAnnotate is an automated annotation tool built using OpenCV. It is a simple tool that is designed to help users label and annotate images and videos using computer vision techniques.
Building an automated image annotation tool using basic OpenCV algorithms. Colorspace, thresholding, and contour analysis. Annotate single class objects easily.
Find out what is EXIF and how EXIF data can be used in various applications. Using EXIF to visualize potholes of a city on google maps.
CenterNet: Object as Points is one of the milestones in the anchor-free (anchorless) object detection algorithm. Anchor-free object detection is more generalizable in other computer vision tasks, e.g., pose estimation,
