ICCV2023 paper list

ICCV2023结果陆续都出来了，收到了很多朋友中稿的消息，ICCV 2023今年一共收录 2100多篇，自动驾驶之心也第一时间进行了跟进，将已确定中稿的工作分享给大家，后面将会持续更新！

后面将会按照3D目标检测、BEV、协同感知、语义分割、点云、SLAM、大模型、NeRF、端到端、多模态融合等多个方向罗列！

如果您的工作也需要被收录，欢迎提交Issue,或者联系邮箱autodrivingtech@163.com，我们会及时收录！

本内容由自公众号【自动驾驶之心】 团队整理，自动驾驶之心建立了一系列技术交流群，面向自动驾驶与AI领域，包括：目标检测、语义分割、全景分割、实例分割、车道线、目标跟踪、3D目标检测、多模态感知、BEV感知、Occupancy、多传感器融合、多传感器标定、transformer、大模型、点云处理、端到端自动驾驶、SLAM、光流估计、深度估计、轨迹预测、高精地图、NeRF、规划控制、模型部署落地、自动驾驶仿真测试、产品经理、硬件配置、AI求职交流等！

如果您有需要，欢迎加入自动驾驶之心：技术交流群

1）OCC感知

SurroundOcc: Multi-Camera 3D Occupancy Prediction for Autonomous Driving

• Paper：https://arxiv.org/abs/2303.09551
• Code：https://github.com/weiyithu/SurroundOcc

OccNet: Scene as Occupancy

• Paper：https://arxiv.org/pdf/2306.02851.pdf
• Code：https://github.com/OpenDriveLab/OccNet

OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy Prediction

• Paper: https://arxiv.org/pdf/2304.05316.pdf
• Code: https://github.com/zhangyp15/OccFormer

OpenOccupancy: A Large Scale Benchmark for Surrounding Semantic Occupancy Perception

• Paper: https://arxiv.org/pdf/2303.03991.pdf
• Code: https://github.com/JeffWang987/OpenOccupancy

2) 端到端自动驾驶

VAD: Vectorized Scene Representation for Efficient Autonomous Driving

• Paper: https://arxiv.org/pdf/2303.12077.pdf
• Code: https://github.com/hustvl/VAD

DriveAdapter: New Paradigm for End-to-End Autonomous Driving to Alleviate Causal Confusion

• Paper: https://arxiv.org/pdf/2308.00398.pdf
• Code: https://github.com/OpenDriveLab/DriveAdapter

3）协同感知

Among Us: Adversarially Robust Collaborative Perception by Consensus

• Paper: https://arxiv.org/pdf/2303.09495.pdf
• Code: https://github.com/coperception/ROBOSAC

HM-ViT: Hetero-modal Vehicle-to-Vehicle Cooperative perception with vision transformer

• Paper: https://arxiv.org/pdf/2304.10628.pdf

Optimizing the Placement of Roadside LiDARs for Autonomous Driving

待更新！

UMC: A Unified Bandwidth-efficient and Multi-resolution based Collaborative Perception Framework

• Paper: https://arxiv.org/pdf/2303.12400.pdf

ADAPT: Efficient Multi-Agent Trajectory Prediction with Adaptation

• Paper: https://arxiv.org/pdf/2307.14187.pdf
• Code: https://github.com/KUIS-AI/adapt

CORE: Cooperative Reconstruction for Multi-Agent Perception

• Paper: https://arxiv.org/pdf/2307.11514.pdf
• Code: https://github.com/zllxot/CORE

4）3D目标检测

PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images

• Paper: https://arxiv.org/abs/2206.01256
• Code: https://github.com/megvii-research/PETR

StreamPETR: Exploring Object-Centric Temporal Modeling for Efficient Multi-View 3D Object Detection

• Paper: https://arxiv.org/pdf/2303.11926.pdf
• Code: https://github.com/exiawsh/StreamPETR.git

Cross Modal Transformer: Towards Fast and Robust 3D Object Detection

• Paper: https://arxiv.org/pdf/2301.01283.pdf
• Code: https://github.com/junjie18/CMT.git

DQS3D: Densely-matched Quantization-aware Semi-supervised 3D Detection

• Paper: https://arxiv.org/abs/2304.13031
• Code: https://github.com/AIR-DISCOVER/DQS3D

SparseFusion: Fusing Multi-Modal Sparse Representations for Multi-Sensor 3D Object Detection

• Paper: https://arxiv.org/abs/2304.14340
• Code: https://github.com/yichen928/SparseFusion

MetaBEV: Solving Sensor Failures for BEV Detection and Map Segmentation

• Paper: https://arxiv.org/pdf/2304.09801.pdf
• Code: https://github.com/ChongjianGE/MetaBEV

Temporal Enhanced Training of Multi-view 3D Object Detector via Historical Object Prediction

• Paper: https://arxiv.org/pdf/2304.00967.pdf
• Code: https://github.com/Sense-X/HoP

Revisiting Domain-Adaptive 3D Object Detection by Reliable, Diverse and Class-balanced Pseudo-Labeling

• Paper: https://arxiv.org/pdf/2307.07944.pdf
• Code: https://github.com/zhuoxiao-chen/ReDB-DA-3Ddet

Learning from Noisy Data for Semi-Supervised 3D Object Detection

• Paper: 待更新！
• Code: https://github.com/zehuichen123/NoiseDet

SA-BEV: Generating Semantic-Aware Bird's-Eye-View Feature for Multi-view 3D Object Detection

• Paper: https://arxiv.org/pdf/2307.11477.pdf

• Code: https://github.com/mengtan00/SA-BEV

PG-RCNN: Semantic Surface Point Generation for 3D Object Detection

• Paper: https://arxiv.org/pdf/2307.12637.pdf
• Code: https://github.com/quotation2520/PG-RCNN

5）语义分割

Rethinking Range View Representation for LiDAR Segmentation

Paper：https://arxiv.org/pdf/2303.05367.pdf

UniSeg: A Unified Multi-Modal LiDAR Segmentation Network and the OpenPCSeg Codebase

已收录，arxiv上暂未放出！

Segment Anything

• Paper: https://arxiv.org/abs/2304.02643
• Code: https://github.com/facebookresearch/segment-anything

MARS: Model-agnostic Biased Object Removal without Additional Supervision for Weakly-Supervised Semantic Segmentation

• Paper: https://arxiv.org/abs/2304.09913
• Code: https://github.com/shjo-april/MARS

Tube-Link: A Flexible Cross Tube Baseline for Universal Video Segmentation

• Paper: https://arxiv.org/pdf/2303.12782.pdf

• Code: https://github.com/lxtGH/Tube-Link

CPCM: Contextual Point Cloud Modeling for Weakly-supervised Point Cloud Semantic Segmentation

• Paper: https://arxiv.org/pdf/2307.10316.pdf
• Code: https://github.com/lizhaoliu-Lec/CPCM

To Adapt or Not to Adapt? Real-Time Adaptation for Semantic Segmentation

• Paper: https://arxiv.org/pdf/2307.15063.pdf
• Code: https://github.com/MarcBotet/hamlet

PointDC: Unsupervised Semantic Segmentation of 3D Point Clouds via Cross-modal Distillation and Super-Voxel Clustering

• Paper: https://arxiv.org/abs/2304.08965
• Code: https://github.com/HalvesChen/PointDC

Contrastive Model Adaptation for Cross-Condition Robustness in Semantic Segmentation

• Paper: https://arxiv.org/pdf/2303.05194.pdf
• Code: https://github.com/brdav/cma

PODA: Prompt-driven Zero-shot Domain Adaptation

• Paper: https://arxiv.org/pdf/2212.03241.pdf
• Code: https://github.com/astra-vision/PODA

Similarity Min-Max: Zero-Shot Day-Night Domain Adaptation

• Paper: https://red-fairy.github.io/ZeroShotDayNightDA-Webpage/paper.pdf
• Code: https://github.com/Red-Fairy/ZeroShotDayNightDA

6）点云感知

Robo3D: Towards Robust and Reliable 3D Perception against Corruptions

• Paper：https://arxiv.org/pdf/2303.17597.pdf
• Code：https://github.com/ldkong1205/Robo3D

Implicit Autoencoder for Point Cloud Self-supervised Representation Learning

• Paper: https://arxiv.org/pdf/2201.00785.pdf
• Code: https://github.com/SimingYan/IAE

P2C: Self-Supervised Point Cloud Completion from Single Partial Clouds

• Paper:
• Code: https://github.com/CuiRuikai/Partial2Complete

CLIP2Point: Transfer CLIP to Point Cloud Classification with Image-Depth Pre-training

• Paper: https://arxiv.org/pdf/2210.01055.pdf
• Code: https://github.com/tyhuang0428/CLIP2Point

SVDFormer: Complementing Point Cloud via Self-view Augmentation and Self-structure Dual-generator

• Paper: https://arxiv.org/pdf/2307.08492.pdf
• Code: https://github.com/czvvd/SVDFormer

AdaptPoint: Sample-adaptive Augmentation for Point Cloud Recognition Against Real-world Corruptions

• Paper: 待更新！
• Code: https://github.com/Roywangj/AdaptPoint/tree/main

RegFormer: An Efficient Projection-Aware Transformer Network for Large-Scale Point Cloud Registration

• Paper: https://arxiv.org/pdf/2303.12384.pdf
• Code: https://github.com/IRMVLab/RegFormer

Point Cloud regression with new algebraical representation on ModelNet40 datasets

• Paper: 待更新！
• Code: https://github.com/flatironinstitute/PointCloud_Regression

Clustering based Point Cloud Representation Learning for 3D Analysis

• Paper: https://arxiv.org/pdf/2307.14605.pdf
• Code: https://github.com/FengZicai/Cluster3Dseg

Implicit Autoencoder for Point Cloud Self-supervised Representation Learning

• Paper: https://arxiv.org/pdf/2201.00785.pdf
• Code: https://github.com/SimingYan/IAE

7）目标跟踪

PVT++: A Simple End-to-End Latency-Aware Visual Tracking Framework

• Paper: https://arxiv.org/pdf/2211.11629.pdf
• Code: https://github.com/Jaraxxus-Me/PVT_pp

Cross-modal Orthogonal High-rank Augmentation for RGB-Event Transformer-trackers

• Paper: 待更新！
• Code: https://github.com/ZHU-Zhiyu/High-Rank_RGB-Event_Tracker

ReST: A Reconfigurable Spatial-Temporal Graph Model for Multi-Camera Multi-Object Tracking

• Paper: 待更新！
• Code: https://github.com/chengche6230/ReST

Multiple Planar Object Tracking

• Paper: 待更新！
• Code: https://github.com/nku-zhichengzhang/MPOT

3DMOTFormer: Graph Transformer for Online 3D Multi-Object Tracking

• Paper: 待更新！
• Code: https://github.com/dsx0511/3DMOTFormer

MBPTrack: Improving 3D Point Cloud Tracking with Memory Networks and Box Priors

• Paper: https://arxiv.org/pdf/2303.05071.pdf
• Code: https://github.com/slothfulxtx/MBPTrack3D

8) 轨迹预测

EigenTrajectory: Low-Rank Descriptors for Multi-Modal Trajectory Forecasting

• Paper: https://arxiv.org/pdf/2307.09306.pdf
• Code: https://github.com/InhwanBae/EigenTrajectory

9）NeRF

IntrinsicNeRF: Learning Intrinsic Neural Radiance Fields for Editable Novel View Synthesis

• Paper: https://arxiv.org/abs/2210.00647
• Code: https://github.com/zju3dv/IntrinsicNeRF

SceneRF: Self-Supervised Monocular 3D Scene Reconstruction with Radiance Fields

• Paper: https://arxiv.org/pdf/2212.02501.pdf
• Code: https://github.com/astra-vision/SceneRF

Single-Stage Diffusion NeRF

• Paper: https://arxiv.org/abs/2304.06714
• Code: https://github.com/Lakonik/SSDNeRF

10）光流

SemARFlow: Injecting Semantics into Unsupervised Optical Flow Estimation for Autonomous Driving

• Paper: https://arxiv.org/pdf/2303.06209.pdf
• Code: https://github.com/duke-vision/semantic-unsup-flow-release

11）双目

ELFNet: Evidential Local-global Fusion for Stereo Matching

• Paper: https://arxiv.org/pdf/2308.00728.pdf
• Code: https://github.com/jimmy19991222/ELFNet

12）鱼眼

SimFIR: A Simple Framework for Fisheye Image Rectification with Self-supervised Representation Learning

• Paper: 待更新
• Code: https://github.com/fh2019ustc/SimFIR