Deepti Hegde's repositories
CLIP-goes-3D
Official code release of "CLIP goes 3D: Leveraging Prompt Tuning for Language Grounded 3D Recognition"
Real-Time-Multiple-Person-Recognition-and-Tracking-for-CCTV-Camera
a surveillance system for CCTV cameras which recognizes selected multiple target individuals and tracks in real time across multiple cameras, with detection, recognition, and kernel-based tracking modules. Facial recognition is done using HOG features and image embedding using OpenFace. We were able to perform simultaneous tracking and recognition of multiple individuals across multiple cameras in real time. Winning project, Smart India Hackathon 2019.
AttentivePrototypeSFUDA
Official implementation for the paper "Attentive Prototypes for Source-free Unsupervised Domain Adaptive 3D Object Detection"
UncertaintyAwareMeanTeacher
The official implementation of the method of Uncertainty-aware Mean Teacher for Source-free Unsupervised Domain Adaptive Object Detection
FairytaleGenerator
An interactive fairytale generator using LSTM networks
APS
Code Repo for Algorithmic Problem Solving Course
CurveNet
Official implementation of "Walk in the Cloud: Learning Curves for Point Clouds Shape Analysis", ICCV 2021
Low-Memory-GEMM-Convolutions
"Memory Efficient GEMM based Convolution Implementation" - Memory footprint reduction for GEMM computation on memory restricted devices.
OpenPCDet
OpenPCDet Toolbox for LiDAR-based 3D Object Detection.
Point-BERT
[CVPR 2022] Pre-Training 3D Point Cloud Transformers with Masked Point Modeling
pyTorchChamferDistance
Implementation of the Chamfer Distance as a module for pyTorch
Relocalisation-of-Agent-In-3D-Map
A lightweight system for camera pose estimation based on hierarchical cluster tree of ORB features
segment-anything
The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.
semantic-segmentation
Nvidia Semantic Segmentation monorepo
ST3D
(CVPR 2021) ST3D: Self-training for Unsupervised Domain Adaptation on 3D Object Detection
vision
Datasets, Transforms and Models specific to Computer Vision