• Stars
    star
    166
  • Rank 227,748 (Top 5 %)
  • Language
    Python
  • Created over 1 year ago
  • Updated about 2 months ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

SSCBench: A Large-Scale 3D Semantic Scene Completion Benchmark for Autonomous Driving

SSCBench: A Large-Scale 3D Semantic Scene Completion Benchmark for Autonomous Driving

Yiming Li*, Sihang Li*, Xinhao Liu*, Moonjun Gong*, Kenan Li, Nuo Chen, Zijun Wang, Zhiheng Li, Tao Jiang, Fisher Yu, Yue Wang, Hang Zhao, Zhiding Yu, Chen Feng

[PDF] [Project]

SSCBench-KITTI-360

SSCBench-nuScenes

SSCBench-Waymo

πŸ”₯[New] SSCBench-PandaSet

News

  • [2023/08]: We add demo for SSCBench-PandaSet. We are working on incoporating the dataset
  • [2023/06]: We release SSCBench-KITTI-360 and SSCBench-nuScenes for academic usage.
  • [2023/06]: SSCBench is submitted to NeurIPS 2023 Track on Datasets and Benchmarks. The preprint version is available on arXiv.

Abstract

Semantic scene completion (SSC) is crucial for holistic 3D scene understanding by jointly estimating semantics and geometry from sparse observations. However, progress in SSC, particularly in autonomous driving scenarios, is hindered by the scarcity of high-quality datasets. To overcome this challenge, we introduce SSCBench, a comprehensive benchmark that integrates scenes from widely-used automotive datasets (e.g., KITTI-360, nuScenes, and Waymo). SSCBench follows an established setup and format in the community, facilitating the easy exploration of the camera- and LiDAR-based SSC across various real-world scenarios. We present quantitative and qualitative evaluations of state-of-the-art algorithms on SSCBench and commit to continuously incorporating novel automotive datasets and SSC algorithms to drive further advancements in this field.

SSCBench Dataset

SSCBench consists of three carefully designed datasets, all based on existing data sources. For more details, please refer to the dataset folder.

Model Checkpoints

We provide the model checkpoints of the experiments reported in the paper. The checkpoints can be accessed on google drive.

Note that the provided checkpoints are trained with the unified class labels. Details of class mappings can be found in the configs folder.

License

Due to the license of the different original datasets, we release SSCBench under the following licenses:

For more details, please refer to the dataset folder file.

Bibtex

If this work is helpful for your research, please cite the following BibTeX entry.

@article{li2023sscbench,
      title={SSCBench: A Large-Scale 3D Semantic Scene Completion Benchmark for Autonomous Driving}, 
      author={Li, Yiming and Li, Sihang and Liu, Xinhao and Gong, Moonjun and Li, Kenan and Chen, Nuo and Wang, Zijun and Li, Zhiheng and Jiang, Tao and Yu, Fisher and Wang, Yue and Zhao, Hang and Yu, Zhiding and Feng, Chen},
      journal={arXiv preprint arXiv:2306.09001},
      year={2023}
}

Star History

Star History Chart

More Repositories

1

DeepMapping

[CVPR2019 Oral] Self-supervised Point Cloud Map Estimation
Python
193
star
2

DeepMapping2

[CVPR2023] DeepMapping2: Self-Supervised Large-Scale LiDAR Map Optimization
Python
156
star
3

DiscoNet

[NeurIPS2021] Learning Distilled Collaboration Graph for Multi-Agent Perception
136
star
4

peac

[ICRA2014] Fast Plane Extraction Using Agglomerative Hierarchical Clustering (AHC)
C++
134
star
5

Occ4cast

Occ4cast: LiDAR-based 4D Occupancy Completion and Forecasting
Python
112
star
6

V2X-Sim

[RA-L2022] V2X-Sim Dataset and Benchmark
111
star
7

FLAT

[ICCV2021 Oral] Fooling LiDAR by Attacking GPS Trajectory
Python
67
star
8

SPARE3D

[CVPR2020] A Dataset for SPAtial REasoning on Three-View Line Drawings
Python
52
star
9

MARS

[CVPR2024] Multiagent Multitraversal Multimodal Self-Driving: Open MARS Dataset
Python
36
star
10

NYU-VPR

[IROS2021] NYU-VPR: Long-Term Visual Place Recognition Benchmark with View Direction and Data Anonymization Influences
Python
31
star
11

EgoPAT3D

[CVPR 2022] Egocentric Action Target Prediction in 3D
Jupyter Notebook
29
star
12

LLM4VPR

Can multimodal LLM help visual place recognition?
Python
28
star
13

insta360_ros_driver

A ROS driver for Insta360 cameras, enabling real-time image capture, processing, and publishing in ROS environments.
Python
27
star
14

DeepExplorer

[RSS2023] Metric-Free Exploration for Topological Mapping by Task and Motion Imitation in Feature Space
Python
26
star
15

Self-Supervised-SPARE3D

[CVPR 2022] Self-supervised Spatial Reasoning on Multi-View Line Drawings
Python
24
star
16

RealCity3D

Jupyter Notebook
22
star
17

DeepSoRo

[RA-L/ICRA2020] Real-time Soft Body 3D Proprioception via Deep Vision-based Sensing
Python
22
star
18

FusionSense

Integrates the vision, touch, and common-sense information of foundational models, customized to the agent's perceptual needs.
Python
19
star
19

SNAC

[ICLR2023] Learning Simultaneous Navigation and Construction in Grid Worlds
Python
18
star
20

SeeDo

Human Demo Videos to Robot Action Plans
Python
16
star
21

TF-VPR

Self-supervised place recognition by exploring temporal and feature neighborhoods
Python
15
star
22

DeepParticleRobot

[ICRA'22] A Deep Reinforcement Learning Environment for Particle Robot Navigation and Object Manipulation
Python
13
star
23

pyAprilTag

python wrapper of AprilTag implemented in cv2cg, used for Robot Perception course
C++
10
star
24

NYC-Indoor-VPR

Python
9
star
25

BAGSFit

Primitive Fitting Using Deep Boundary Aware Geometric Segmentation
9
star
26

vis_nav_player

[ROB-GY 6203] Example Visual Navigation Player Code for Course Project
Python
5
star
27

NYC-Event-VPR

Python
5
star
28

Mobile3DPrinting

https://ai4ce.github.io/Mobile3DPrinting/
4
star
29

PointCloudSimulator

Code for simulating 2D point clouds
Python
3
star
30

M3DP-Sim

C++
2
star
31

pyP2Mesh

python wrapper for finding a point's closest point on a triangle mesh
Python
2
star
32

VIP_SelfDrive

Makefile
2
star
33

EgoPAT3Dv2

Python
2
star
34

LUWA

Jupyter Notebook
1
star
35

ai4ce_robot_ROS2_drivers

This repo contains all the ROS2 driver packages modified at AI4CE lab for working with various robots
1
star