PETR
[ECCV2022] PETR: Position Embedding Transformation for Multi-View 3D Object Detection & [ICCV2023] PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images
AI summary
Position Embedding Framework
Develops a framework for multi-view 3D object detection and perception from camera images using position embedding transformation.
- stars
- 881
- forks
- 132
- watching
- 15
- awesome list
- 1
Similar projects
Found by comparing what the projects do, not just their names.
Localizer
Improves image restoration performance by converting global operations to local ones during inference
3D Vision Framework
A software framework for 3D vision and computer vision tasks using deep learning and 2D keypoints.
Image embedding
An implementation of a deep learning-based image representation learning approach using a modified fully connected layer and transfer learning from VGG16
View transformer
This project focuses on manipulating 3D views using deep learning techniques.
Vector embedding utility
A fast and efficient utility package for utilizing vector embeddings in machine learning models
Region-of-Interest Training
Training and deploying large language models on computer vision tasks using region-of-interest inputs
Perception toolkit
ROS package for perception processing and evaluation in the KITTI Vision Benchmark Suite
Plane detector
An implementation of a deep learning-based method for 3D plane recovery from images.
Change Detection
Automatically inferring 2D and 3D change detection maps from bitemporal optical images without relying on DSMs.
Model generator library
Python bindings for CityEngine's procedural runtime for generating 3D models
Camera Laser Calibration Tool
Automates extrinsic calibration of cameras and 2D lasers in robotics using ROS
Visual Inertial Odometry Library
An implementation of a monocular visual inertial odometry algorithm based on Multi-State Constraint Kalman Filter for accurate and robust localization
Robot Navigation
Develops robot navigation policies in crowded spaces using reinforcement learning and attention mechanisms.
Visual representation generator
A project that generates visual representations tailored for general visual reasoning, leveraging hierarchical scene descriptions and instance-level world knowledge.
Generative Model
This project aims to develop a generative model for 3D multi-object scenes using a novel network architecture inspired by auto-encoding and generative adversarial networks.
Featured in 1 awesome list
Each link jumps to the spot where the list mentions PETR.
