Popular GitHub Repositories for Normal

micronet, a model compression and deploy lib. compression: 1、quantization: quantization-aware-training(QAT), High-Bit(>2b)(DoReFa/Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference)、Low-Bit(≤2b)/Ternary and Binary(TWN/BNN/XNOR-Net); post-training-quantization(PTQ), 8-bit(tensorrt); 2、 pruning: normal、regular and group convolutional channel pruning; 3、 group convolution structure; 4、batch-normalization fuse for quantization. deploy: tensorrt, fp32/fp16/int8(ptq-calibration)、op-adapt(upsample)、dynamic_shape

[CVPR'22] ICON: Implicit Clothed humans Obtained from Normals

[CVPR 2024 Oral] Rethinking Inductive Biases for Surface Normal Estimation

Relation-Shape Convolutional Neural Network for Point Cloud Analysis (CVPR 2019 Oral & Best paper finalist)

[CVPR 2024] Official repository of "Material Palette: Extraction of Materials from a Single Real-world Image"

Stereo vision based object detection for ADAS, autonomous vehicle

Official implementation of CVPR 2024 PromptAD: Learning Prompts with Only Normal Samples for Few-Shot Anomaly Detection

Code underlying our publication "Modeling the Distribution of Normal Data in Pre-Trained Deep Features for Anomaly Detection" at ICPR2020

In this project, I implemented several ensemble methods (including bagging, AdaBoost, SAMME, stacking, snapshot ensemble) for a normal CNN model and Residual Neural Network.

RNb-NeuS: Reflectance and Normal-based Multi-view 3D Reconstruction.

[ECCV 2024] Learning to Detect Multi-class Anomalies with Just One Normal Image Prompt

Self-contained, minimalistic implementation of a language model that generates coherent and normal sounding names. It uses an input dataset of names and probability distribution to generate new names based on the sequences of four characters.

Photoshopped and leaked nude photos of celebrities and normal people are circulated throughout the internet. This project is aimed to report all those images (Using Opencv and imagga/Tensorflow) to the search engine with image URL and authentication id to remove the content.

The basic idea is to utilize a U-Net, modified perceptual loss (pearson instead of MAE), learned basis functions, and "mean teacher" training in order to synthesize images of high quality without the normal troubles of training a GAN.

RNb-NeuS2: Multi-View Surface Reconstruction Using Normal and Reflectance Cues

A project to show howto use SpringAI with OpenAI to chat with the documents in a library. Documents are stored in a normal/vector database. The AI is used to create embeddings from documents that are stored in the vector database. The vector database is used to query for the nearest document. That document is used by the AI to generate the answer.

? Use Bi-normal Separation to find document vectors which is used to compute similarity for shorter sentences.

RankSEG: A consistent ranking-based framework for segmentation

:eyeglasses: An OpenCV project for detecting books, papers, or any rectangular objects in an image, regardless the perspective, and straightening it as if it was a normal scan.

[IJCV 2024] Hard-normal Example-aware Template Mutual Matching for Industrial Anomaly Detection