esp32-cam face recognition with MQTT and esp-who framework

Ultra-lightweight human detection. The number of parameters does not correlate to inference speed. For limited use cases, an input image resolution of 64x64 is sufficient. High-level object detection architectures such as YOLO are overkill.

World's First NMS-Free YOLOv26n on ESP32-P4. Features end-to-end Int8 QAT and custom C++ optimizations achieving 30% faster inference than the official ESP-DL YOLOv11n (1.7s vs 2.4s).

Deploy deep neural network models on esp32 SOC.

基于Ultralytics YOLO11与Espressif ESP-DL的嵌入式 AI 视觉部署全栈方案。

CNN-based auto-exposure control for ESP32-P4-EYE. INT8 quantized model (128x128 RGB) analyzes brightness/contrast/saturation in real-time (15ms inference). Dual-mode: camera native AE vs DL control. Manual preprocessing workaround for ESP-DL bug. OV2710 sensor, 500ms adjustment cycles, Telegram integration.

面向 ESP32-S3 的完整人脸 AI 流程实现。涵盖从原始图像采集、多级人脸对齐到特征匹配的全过程，针对 ESP-DL 深度学习库进行了底层 API 适配与稳定性优化。