Awesome GitHub RepositoriesTensor Processing Pipelines

This project is a collection of pre-trained machine learning models and conversion pipelines designed for running inference directly in the browser using TensorFlow.js. It provides a library of ready-to-use models for computer vision, audio classification, and natural language processing tasks. The suite includes specialized tools for transforming Python-based Keras models into JSON formats compatible with web environments. It enables the deployment of these models by fetching architectures and weight shards via HTTP for client-side execution. The project covers a broad range of capabilities

Final2x is an AI image super-resolution tool and neural network inference engine designed to increase image resolution and reconstruct missing details while reducing noise. It functions as a cross-platform image upscaler that executes consistent super-resolution logic across different operating systems. The project serves as a custom model inference engine and upscaling interface, allowing for the import and application of user-defined super-resolution weights and architectures to tailor the visual output of enlarged images. The system utilizes hardware-accelerated processing to offload comp

Darkflow is an object detection framework and computer vision pipeline that provides a programmatic interface for performing real-time image analysis and object identification. It functions as a tool for loading weights, fine-tuning models, and executing inference on both static images and video feeds. The project serves as a converter that translates Darknet configurations and weights into TensorFlow graphs to enable retraining and deployment. It includes a model exporter that saves trained graphs into portable protobuf files for use on mobile and native devices. The system covers capabilit

Clarity-upscaler 是一款 AI 图像放大与增强工具，利用深度学习模型提高图像分辨率并恢复视觉细节。它作为一个超分辨率推理引擎，使用神经网络预测缺失像素，并从低分辨率源中合成高频细节。 该项目以可编程 API 的形式提供，允许将自动高分辨率图像处理和锐化功能集成到外部应用程序和工作流中。该接口支持通过编程方式放大图像，以创建高分辨率资产。 该系统提供自动图像增强功能，通过去除噪点和提高清晰度，生成低质量图像的更锐利版本。它通过解耦的客户端-服务器架构来处理机器学习推理中计算密集型的任务。

This project is an AI upscaling framework and deep learning image restorer designed to estimate original source pixels from low-resolution inputs. It functions as a super-resolution reconstruction system that transforms pixelated images into high-resolution versions by restoring high-frequency details and sharpening edges. The system utilizes a convolutional neural network pipeline to analyze pixel data and perform digital image restoration. It employs pixel-shuffle upsampling to rearrange channel dimensions into spatial dimensions, which increases resolution while reducing checkerboard artif

RestorePhotos 是一款 AI 面部修复工具和深度学习图像超分辨率工具，旨在消除模糊并重建退化面部照片中丢失的细节。它作为面部照片增强器和生成对抗网络（GAN）图像处理器，将低质量像素转换为高分辨率的面部特征。 该系统利用 GPU 加速的推理引擎来运行机器学习模型，进行实时图像修复。这种硬件加速支持锐化面部图像和提高视觉保真度所需的繁重矩阵乘法和基于张量的运算。 该项目涵盖了 AI 照片修复和数字档案修复，特别专注于面部图像增强。它采用顺序处理管道来执行深度学习图像超分辨率，通过重建缺失的细节来提高图像的分辨率和清晰度。

该项目是一个 ComfyUI ControlNet 预处理器套件和计算机视觉分析工具包。它作为一个稳定扩散（Stable Diffusion）图像预处理器，从图像中提取结构提示以引导潜在扩散工作流。 该系统提供用于人体姿态估计（包括骨骼关键点和面部网格）以及通过深度和表面法线估计进行 3D 场景映射的专用模型。它还包括使用光流分析进行 AI 视频运动控制的工具。 更广泛的能力范围涵盖图像结构分析（如线条艺术、边缘提取和语义分割），以及亮度隔离和颜色及风格提示生成。这些过程由硬件加速运行时和模型检查点缓存支持，以减少推理延迟。

这是一个使用 TensorFlow 框架构建的生成模型架构库。它提供了用于生成合成数据和逼真图像的实现，特别关注变分自编码器（VAE）和各种生成对抗网络（GAN）变体。 该集合包括特定的 GAN 架构，如 WGAN-GP、LSGAN、InfoGAN 和 EBGAN。它还具有旨在学习潜在表示并从学习到的分布中合成新样本的变分自编码器。 该项目涵盖了用于归一化和裁剪数据的图像处理管道，以及包括 2D 卷积和线性变换在内的神经网络层工具包。它提供了潜在空间操作、条件向量注入以及通过特定损失函数进行模型稳定的功能。

pytorch-fid is a PyTorch-based evaluator and image distribution analysis library used to calculate the Fréchet Inception Distance. It functions as a benchmarking tool that maps image pixels to high-dimensional feature vectors using a pre-trained convolutional neural network to measure the mathematical divergence between real and synthetic datasets. The library quantifies the quality and diversity of generative models by representing image feature sets as mean and covariance matrices. It allows for the extraction of latent representations from specific neural network layers, with configurable

This project is a research-oriented PyTorch framework designed for the implementation and training of generative video diffusion models. It provides a modular toolkit that extends standard image-based diffusion techniques into three dimensions, enabling the synthesis of coherent video sequences through iterative denoising processes. The framework distinguishes itself by utilizing factored space-time attention, which decomposes high-dimensional video data into separate spatial and temporal layers to maintain motion consistency while managing computational complexity. It supports multi-modal tr