Building on an earlier blog, we continue exploring SoccerNet 2026 with the Player-Centric Ball Action Spotting challenge and its winning solution, PAVE.
What changes when a horizontal video becomes vertical? We tested the open-source reframing tools, then built a crop planner around predicted visual attention. An exploration of framing, existing tools, and the choices behind an automatic crop.
We took the winning solution, DENSER, and ran it ourselves. In this post we go through how it works, and answer three questions along the way. How is the data built? How many cameras are really needed? And could this be done with a real football match?
Curious about AI agents? We explore what they are, how to use them, platforms for building them, and some limitations to consider. Discover how these AI systems are simplifying complex tasks by turning natural language into action!
We’re testing a locally deployed Llama-3.2 model to generate UEFA Champions League match reports from play-by-play commentary. Starting with a pre-trained 3B model in ollama, we’ll apply LoRA fine-tuning with unsloth and compare performance, hardware needs, and key tuning parameters.
In this post, we explore fine-tuning GPT-4o for generating authentic, engaging football match reports from play-by-play commentary. By retraining the model on a dataset of UEFA Champions League matches, the goal was to craft outputs that mimic professional sports journalism, with rich context, expressive language, and key insights. The process highlighted the strengths of fine-tuning for achieving consistent, domain-specific results, while balancing cost and performance.
Dive into the future of digital interaction with our comprehensive overview of Text-To-Speech advancement. On this article, we explore how TTS is revolutionizing voice cloning with near-human accuracy, multilingual support, and real-time generation. Besides, we delve into XTTS v2's innovative features, including zero-shot voice cloning and fine-tuning, and see how it stacks up against other leading models.
The EU's journey towards AI regulation culminated in the political agreement on the AI Act in December 2023. Here’s a comprehensive overview of this AI Act.
Build a real-time streaming pipeline with Apache Kafka and Amazon MSK. Learn how to generate, ingest, and process streaming data using Python and Kafka. Enable real-time processing, machine learning integration, and intelligent decision-making in your applications.
Exploring LangChain, an open-source project that is revolutionizing the capabilities of language models. It offers seamless model switching, simplified prompt management, interactive capabilities, memory enhancement, and efficient knowledge extraction. With its versatility and growing community support, LangChain it is a game-changer for language model-powered applications.
On this article, you will find an overview of the new features and improvements introduced in Pandas 2.0, that can greatly improve data processing, analysis, and visualization workflows
Exploring the possibility of using ChatGPT or DaVinci as a sales assistant chatbot on an eCommerce store, without the API for external integration.
On this article you will learn how to generate a 3D reconstruction from a set of images using Meshroom, and make real-world measurements with it
A review of what was unveiled at this year’s NVIDIA GPU Technology Conference, and what was said about Deep Learning and AI during its sessions
From my experience with using the Detectron2 library for object detection and training machine learning models, here's an explanation on how to split a dataset into test, train, and validation sets, register metadata for each, and monitor accuracy on validation while training.
A quick view to the available solutions for automatic ball detection
Following up on our previous article about Hand Detection, this is the second part of our series about the possibilities of hand recognition technology. Here, we dive the design, training, and testing of a Hand Keypoint detector, a Neural Network capable of detecting and tracking hand movements.
Exploring how to extract a complete 3D human body pose from just a single image using SMPLify-X, a revolutionary method to generate a complex, expressive 3D model of the human body, including hands, face, and body
Introducing our series about hand recognition technology, we define the importance of hand recognition, its potential to revolutionize human-device interactions, the evolution of hand recognition technology and the different stages of hand detection and gesture recognition.
