ESP-DL Project Organization

ESP-DL’s modular design enables efficient development, maintenance, and scalability. The project is organized as follows:

dl (Deep Learning)

Core deep learning modules and tools, divided into submodules:

model Loads, manages, and allocates memory for deep learning models. Includes dl_model_base and dl_memory_manager.
module Interfaces for operations (convolution, pooling, activation, etc.). Files: dl_module_conv, dl_module_pool, dl_module_relu etc.
base Implements operations for chips (esp32s3/esp32p4) with assembly support.
math Mathematical operations (matrix functions). Files: dl_math.hpp and dl_math.cpp.
tool Auxiliary functions (utility tools). Files: dl_tool.hpp and dl_tool.cpp.
tensor Tensor classes. Files: dl_tensor_base.hpp.

Computer vision modules divided into submodules:

classification Image classification (model inference). Inference: dl_cls_base. Post-processors: imagenet_cls_postprocessor, dl_cls_postprocessor.
recognition Feature extract (model inference). Feature database management (Enroll, delete, query). Pre-processor: dl_feat_image_preprocessor. Inference: dl_feat_base. Post-processor: dl_feat_postprocessor. Database: dl_recognition_database
image Image process(resize, crop, warp affine). Color conversion(pixel, img). Image preprocessor (pipeline of resize, crop, color conversion, normalization, quantization). Image decoding/encoding (JPEG/BMP). Draw utility (point, hollow rectangle).

Image process: dl_image_process. Color conversion: dl_image_color. Image preprocessor: dl_image_preprocessor. Image decoding/encoding: dl_image_jpeg, dl_image_bmp. Draw utility: dl_image_draw.
detect Object detection (model inference). Inference: dl_detect_base. Post-processors: dl_detect_msr_postprocessor, dl_detect_mnp_postprocessor, dl_detect_pico_postprocessor.

Handles FlatBuffers models: