This article presents a complete citrus detection system using YOLOv8, FastAPI, and Vue3, featuring image/video detection, ablation experiments, negative sample training, and full deployment instructions for agricultural AI applications.
FFmpeg 9.0, codenamed “Lei,” was released on August 4 to commemorate Chinese audio‑video developer Lei Xiaohua, whose contributions to the library and early AI‑related tutorials shaped the ecosystem, while the article also traces FFmpeg’s history, naming conventions, and expanding role in modern AI pipelines.
This article details the pure‑Java mica‑ppocr library built on Java 17 and ONNX Runtime, covering its rapid version evolution, core engine upgrades, structured parsing of six document types, Spring Boot and Solon integration, performance optimizations, and architectural refactors, providing a comprehensive technical overview for developers.
The article presents mica-ppocr, a Java‑17 native implementation of PP‑OCRv6 that eliminates Python and Paddle dependencies, offers three model tiers, Spring Boot starter support, and detailed usage examples, enabling Java back‑ends to perform high‑quality OCR locally and compliantly.
FFmpeg 9.0, codenamed “Lei” to honor Chinese audio‑video developer Lei Xiaohua on the tenth anniversary of his death, continues the project’s evolution from a simple codec suite to a ubiquitous multimedia backbone powering video editing, streaming, and AI pipelines, while tracing its rich history and community contributions.
This article reviews the most capable open‑source Chinese text‑to‑speech models that run entirely on CPU in 2026, compares their quantization and speed features, recommends acceleration engines, outlines five hard‑won optimization rules, and provides a concise selection guide for various deployment scenarios.
This guide walks you through setting up an Ubuntu environment, training a LightGBM stock‑movement model in Python, exporting it to ONNX, and deploying fast, cross‑platform inference in Go using ONNX Runtime, complete with code snippets and project structure.
In this detailed engineering guide we diagnose a heavyweight ERNIE‑Base text‑classification service consuming 128 CPU cores and 96 GB RAM, then apply a three‑step optimization—model selection, structured pruning with PaddleSlim, and engine migration to ONNX Runtime—achieving a 75% reduction in resource usage while keeping recall above 99.5% and boosting inference speed by over 20%.
This article walks through exporting a PyTorch image‑classification model to ONNX, identifying a hidden layer node, adding it as an output, and using ONNX Runtime to retrieve both the final prediction and the intermediate feature tensor.
