flyingGH's repositories
3D-Position-Reconstruction-Using-4-Cams
College project for Computacional Vision that maps the position of a moving robot with an ArUco based on a video of 4 cameras (with known extrinsic and intrinsic parameters)
3D-Semantic-Segmentation-From-2D-Images
Transforming 2D images into 3D semantically segmented scenes using innovative CNN architecture and COLMAP reconstruction.
3d_recon
Computer vision's final homework
autoware-lgsvl-bridge
This is a bridge between autoware.ai and LGSVL simulator
BerfScene
[CVPR 2024] BerfScene: Bev-conditioned Equivariant Radiance Fields for Infinite 3D Scene Generation
BEVFormer
[ECCV 2022] This is the official implementation of BEVFormer, a camera-only framework for autonomous driving perception, e.g., 3D object detection and semantic map segmentation.
Birds-Eye-View-Trajectory-Prediction-for-Autonomous-Driving
This repository contains our work on a comprehensive investigation on motion prediction for Autonomous Vehicles using the PowerBEV framework and a Multi-Camera setup. Validated trajectory forecasting capabilities on the NuScenes, Woven and Argoverse datasets and identified challenges in model generalization across these datasets.
BroadBEV
BroadBEV: Collaborative LiDAR-camera Fusion for Broad-sighted Bird's Eye View Map Construction, in 2024 ICRA.
gaustudio
A Modular Framework for 3D Gaussian Splatting and Beyond
gtsfm
End-to-end SFM framework based on GTSAM
LimSim
LimSim & LimSim++: Integrated traffic and autonomous driving simulators with (M)LLM support
llama3.np
llama3.np is pure NumPy implementation for Llama 3 model.
marepo
[CVPR 2024 Highlight] Map-Relative Pose Regression for Visual Re-Localization
MotionBEVpp
A Moving Object Semantic Segmentation Model Based on the Bird's eye view - jxLiang's Bachelor Thesis Code
python_camera_library
A simple Python based pinhole camera library that implements forward- and back-projection, 3D reconstruction etc.
pytube
A lightweight, dependency-free Python library (and command-line utility) for downloading YouTube Videos.
RGBD-3DGS-SLAM
RGBD-3DGS-SLAM is a monocular SLAM system leveraging 3D Gaussian Splatting (3DGS) for accurate point cloud and visual odometry estimation. By integrating neural networks, it estimates depth and camera intrinsics from RGB images alone, with optional support for additional camera information and depth maps.
Sphere_SFM_Scripts
Run Automation of Sphere_SFM
Structure-from-Motion-SFM--1
3D reconstruction of a scene using multi-view calibrated images by following a Structure from Motion pipeline
uq_bevss
Evaluating Uncertainty Quantification for Bird’s Eye View Semantic Segmentation
workstation-sfm-setup
Dense SFM setup for the workstation