The research project based on Semantic KITTTI dataset, 3d Point Cloud Segmentation , Obstacle Detection

This GitHub repository has been created for the research project titled "Improving Aerial Targeting Precision: A Study on Point Cloud Semantic Segmentation with Advanced Deep Learning Algorithms."

[IJCV] PointSegBase: a codebase for point cloud segmentation research. Related works: Deep Hierarchical Learning for 3D Semantic Segmentation

Developed a Software for semantic segmentation of remote sensing imagery using Fully Convolutional Networks (FCNs). Initially, this software developed for extracting the road network from high-resolution remote sensing imagery. And now, this software can be used to extract various features (Semantic segmentation of remote sensing imagery). This project can also extract from Vector Data. This is part of my Internship at ISRO (Indian SPace Research Organization)'s NRSC (National Remote Sensing Centre) campus.

AgriSegment — A multi-modal plant segmentation suite for agricultural research. It offers four FastAPI-powered tools—Hybrid (SegFormer + SAM), Interactive (real-time SAM), Semantic (batch SegFormer), and Panoptic (Mask2Former)—making AI-driven plant segmentation accessible to non-programmers.

A repository containing code for a class research project in which I investigated the applicability of sparse data fusion between lidar point clouds and RGB imagery for semantic scene segmentation.

In this research, a novel approach has been presented for multi-label semantic segmentation of the martian terrain for navigation using variants of a transformer model - SegFormer, which has resulted in a commendable accuracy of 90.86% and mIoU of 83.55% on the AI4Mars dataset

ACM Research 2021 | Improving the Utility of Temporal LIDAR Data in Semantic Segmentation

Recent work in the field of Explainable AI and Computer Vision on CNN based architecture has improved the interpretability of Deep Learning models and helped in visualizing the model pre- dictions. Methods like CAM, Grad-CAM and Guided Grad-CAM have proved the practicality of localized visual attention in the classification and categorization applications. However, not much research has been done on generative models. In our work, we implement Grad-CAM technique on VAE and CVAE models trained on CelebA-HQ dataset and calculate neural attention map. The aim of the project is to build generative models capable of generating controllable human faces and build semantic segmentation of human face, and then investigate methodologies to improve the explainability by applying Explainable AI tech- niques like Grad-CAM, and analysing the effect of altering the model architecture, loss functions, latent space size. Furthermore, we investigate the latent space information of models by modifying the latent node variables.

PhD Research: Semantic Segmentation on Remotely Sensed Images Using Deep Convolutional Encoder-Decoder Neural Network

This repository contains iPython notebook files that address the problem of breast cancer detection on ultrasound images. The problem of breast cancer detection is approached from several different ways- Benign vs. Malignant BUS Tumor classification(with and without ground truth tumor masks), Semantic Segmentation of Breast Tumor in BUS images, and Multitask learning based segmentation and classification of BUS tumors. The code and the results present in this repository can serve as a baseline for innovative research works in this area.

Semantic Image Segmentation. Paper: https://research.googleblog.com/2018/03/semantic-image-segmentation-with.html

The entire project is programmed in Python. Including how to get Baidu map street view, how to do semantic segmentation (pre-trained model based on Openvino), and how to calculate the green view rate

Pytorch implementation of "CAENet: Efficient Multi-task Learning for Joint Semantic Segmentation and Depth Estimation", accepted in the Research Track in ECML PKDD 2023.

Here are some valuable papers, websites or articles about semantic image segmentation with focus on weak supervision circumstances. Some other materials related to the research field of computer vision and machine learning may also be included.

List of algorithms (papers with open source code) researched for object detection (bounding box, semantic segmentation, instance segmentation)

Our research for AfaMamba: Adaptive Feature Aggregation with Visual State Space Model for Remote Sensing Images Semantic Segmentation

This repository provides a tool to calculate the area of polygons in semantic segmentation data, specifically for images in YOLO format. It enables the precise calculation of areas for objects defined as polygons within images. This tool is useful for researchers and developers in analyzing and evaluating segmentation tasks.

Track of the experiments during my research on Semantic Segmentation

semantic segmentation demo bases on https://github.com/lizhengwei1992/models/tree/master/research/deeplab

To implement “Adversarial Examples for Semantic Segmentation and Object Detection” and “DAPAS : Denoising Autoencoder to Prevent Adversarial attack in Semantic Segmentation” research papers and analyzing results from both implementations.

DeepLab is a semantic segmentation Deep Learning algorithm. Implementation is copied from tenserflow/../research/deeplab on GitHub.

Research internship Project on Detection and Semantic Segmentation of apples

Research code for semantic segmentation of 3D plant point clouds using geometric deep learning, developed in collaboration with Oak Ridge National Laboratory.

Repository for researching, training, testing and inference of fast semantic segmentation models to be used with CPU, on mobile devices or on minicomputers

This repository contains code (research only, no production-ready) for a methodology for semantic segmentation for radar sounder data based on random walks and label propagations, published on IEEE TGRS.

CytoSegment is a Python library for semantic segmentation of cytometry image data using deep learning. It is designed to support researchers and practitioners working with microscopic biological imagery—especially blood cells

The sKwanda_V2: OpenGeoAI-Dataset is a high-resolution optical imagery dataset, designed to support advanced research in geospatial machine learning applications, such as semantic segmentation, land cover classification, and spatial analysis.

A PyTorch implementation of the DeepLab-v3+ model under development. This is basically a subset of a clone of the pytorch-deeplab-xception repo authored by @jfzhang95. This repo is intended for further research on instance-level semantic segmentation.

这是一个关于语义分割的程序代码，常用于街景影像的研究使用 This is a program code for semantic segmentation, commonly used in street view image research.