65 skills found · Page 3 of 3
thealonemusk / LD Net Lightweight Dehazing NetworkLD-Net is a deep learning-based algorithm designed to enhance visibility in hazy images and videos, improving visual quality and clarity. This project implements a lightweight dehazing network based on a convolutional autoencoder (CAE) architecture.
Toandinh1 / HPE Li ECCV2024This repo is the official implementation for HPE-Li: WiFi-enabled Lightweight Dual Selective Kernel Convolution for Human Pose Estimation (ECCV 2024).
Phil-y / PMS NetPMS-Net: A Lightweight Convolutional Attention U-Shaped Network for Medical Image Segmentation
guillaume-chevalier / Python Conv LibA lightweight library to do for-loop-styled convolution passes on your iterable objects (e.g.: on a list). Note: this is not a convolution, it is about exposing what would the kernel pass on in the first place in your loops.
Ali5hadman / PAL Net A Point Wise CNN With Patch AttentionPAL-Net is a lightweight neural network for automatic 3D facial landmark localization using point-wise convolutions and a patch-attention mechanism. This repository includes the full training and evaluation pipeline with support for custom data.
DerOzean / Cable Detection In Automated Dissassembly Environment Using Deep LearningCables are essential components of any device in the electronic waste sector. For several years disassembly processes in this sector got automated step by step. However, most of these automation’s were device-specific and not usable on a wide range 1 . An advantage came about as neural networks improved in image recognition. The recent improvements marked the starting point for many research groups to focus on disassembly lines that can handle multiple devices. This bachelor thesis aims to address a fundamental problem in such disassembly processes: Cable-Detection . Consider the task of disassembling a DVD-Player which still includes usable components. Cables connect most of these components for information and energy exchange. Therefore one main goal before removing individual components is to cut the cables between the components to ensure their safe removal. Inevitably this task needs a precise knowledge about the position of every single cable. For a disassembly robot, a strategy that is worth pursuing is to locate the cables process- ing an RGB image. An RGB image is a lightweight solution with high standards regarding the improvement in camera technology over the last year through mobile devices. This Bachelor thesis presents a model for cable detection. A state-of-the-art instance segmentation model Mask R-CNN 2 , published in 2018, is trained. The presented work is public on GitHub to make this global concerning topic available for the public. Keywords: Deep Convolutional Neuronal Networks, Mask R-CNN, Cable-Detection, Disassembly, E-Waste
matteorisso / PIT[DAC 2021] Pruning In Time (PIT): A Lightweight Network Architecture Optimizer for Temporal Convolutional Networks
elcruzo / Cuda ConvLightweight CUDA kernel for 2D image convolution achieving 20x+ speedup. Built with CuPy.
Bo-Zhou-gogogo / RamanNetCode for the paper:"RamanNet: A lightweight convolutional neural network for bacterial identification based on Raman spectra "
asemorales / MobilelooknetLightweight neural network for cancer detection in bone scans using PyTorch. Code repository for the paper "A Lightweight Convolutional Neural Network for Detection of Osseous Metastasis using Feature Fusion and Attention Strategies". Presented at CVIPPR 2024.
IDU-CVLab / COV19DICCV 2021 Workshop: MIA-COV19D : A lightweight automated solution for COVID-19 diagnosis using image processing techniques and Convolutional Neural Networks
WouterDurnez / AirheadLightweight Convolutional Neural Networks for Brain Tumor Segmentation
princeSmall / CoreML MobileNetMobileNets are based on a streamlined architecture that have depth-wise separable convolutions to build lightweight, deep neural networks.
BCJuan / BronchoTrackLightweight bronchoscopy tracking through a hierarchically pruned and distilled recurrent convolutional neural network
Geo3D-AI-CSU / 3DCNN4MPMA lightweight convolutional neural network with end-to-end learning for three-dimensional mineral prospectivity modeling
LucaHermes / Lightweight Motion ForecastingA lightweight graph-convolutional model for skeletal human motion forecasting on the Human3.6M (H3.6M) dataset.
xuke172627902 / MSANmodels and codes for Multi-scale strip-shaped convolution attention network for lightweight image super-resolution (MSAN)
dyl96 / LRDNetA Lightweight Road Detection Algorithm Based on Multiscale Convolutional Attention Network and Coupled Decoder Head (IEEE GRSL 2023)
jemin-023 / Tiny Cnn In CppA lightweight, dependency-free Convolutional Neural Network (CNN) engine written in C++17. This project focuses on implementing the fundamental mathematics of deep learning—convolution, backpropagation, and gradient descent—directly on the CPU.
Rh-Dang / CA MSN Action RecognitionBased on the skeleton action recognition of graph network, the optimization was carried out from the aspects of graph convolution lightweight, channel attention mechanism and multi-scale time feature aggregation.
