ECCV 2024
自动驾驶 · 具身智能论文归档

Making Large Language Models Better Planners with Reasoning-Decision Alignment

DriveLM: Driving with Graph Visual Question Answering

Mask2Map: Vectorized HD Map Construction Using Bird's Eye View Segmentation Masks

MapTracker: Tracking with Strided Memory Fusion for Consistent Vector HD Mapping

Rethinking Data Augmentation for Robust LiDAR Semantic Segmentation in Adverse Weather

SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow

RAPiD-Seg: Range-Aware Pointwise Distance Distribution Networks for 3D LiDAR Segmentation

Bi-directional Contextual Attention for 3D Dense Captioning

PointLLM: Empowering Large Language Models to Understand Point Clouds

RISurConv: Rotation Invariant Surface Attention-Augmented Convolutions for 3D Point Cloud Classification and Segmentation

Projecting Points to Axes: Oriented Object Detection via Point-Axis Representation

LiveHPS++: Robust and Coherent Motion Capture in Dynamic Free Environment

Integer-Valued Training and Spike-driven Inference Spiking Neural Network for High-performance and Energy-efficient Object Detection

Relation DETR: Exploring Explicit Position Relation Prior for Object Detection

WPS-SAM: Towards Weakly-Supervised Part Segmentation with Foundation Models

Align before Collaborate: Mitigating Feature Misalignment for Robust Multi-Agent Perception

M^2Depth: Self-supervised Two-Frame Multi-camera Metric Depth Estimation

OP-Align: Object-level and Part-level Alignment for Self-supervised Category-level Articulated Object Pose Estimation

On Calibration of Object Detectors: Pitfalls, Evaluation and Baselines

Efficient Active Domain Adaptation for Semantic Segmentation by Selecting Information-rich Superpixels

KeypointDETR: An End-to-End 3D Keypoint Detector

HGL: Hierarchical Geometry Learning for Test-time Adaptation in 3D Point Cloud Segmentation

Exploring the Feature Extraction and Relation Modeling For Light-Weight Transformer Tracking

Physics-Based Interaction with 3D Objects via Video Generation

Towards Scene Graph Anticipation

Controlling the World by Sleight of Hand

SemGrasp: Semantic Grasp Generation via Language Aligned Discretization

LEGO: Learning EGOcentric Action Frame Generation via Visual Instruction Tuning

Language-Driven 6-DoF Grasp Detection Using Negative Prompt Guidance

NL2Contact: Natural Language Guided 3D Hand-Object Contact Modeling with Diffusion Model

UGG: Unified Generative Grasping

MaxMI: A Maximal Mutual Information Criterion for Manipulation Concept Discovery

Beyond the Contact: Discovering Comprehensive Affordance for 3D Objects from Pre-trained 2D Diffusion Models

NeRMo: Learning Implicit Neural Representations for 3D Human Motion Prediction

Controllable Human-Object Interaction Synthesis

Propose, Assess, Search: Harnessing LLMs for Goal-Oriented Planning in Instructional Videos

Pose-Aware Self-Supervised Learning with Viewpoint Trajectory Regularization

Correspondences of the Third Kind: Camera Pose Estimation from Object Reflection

A Direct Approach to Viewing Graph Solvability

Camera Calibration using a Collimator System

COMO: Compact Mapping and Odometry

Scene Coordinate Reconstruction: Posing of Image Collections via Incremental Learning of a Relocalizer

ADen: Adaptive Density Representations for Sparse-view Camera Pose Estimation

SPVLoc: Semantic Panoramic Viewport Matching for 6D Camera Localization in Unseen Environments

DVLO: Deep Visual-LiDAR Odometry with Local-to-Global Feature Fusion and Bi-Directional Structure Alignment

Six-Point Method for Multi-Camera Systems with Reduced Solution Space

Grounding Image Matching in 3D with MASt3R

Equi-GSPR: Equivariant SE(3) Graph Network Model for Sparse Point Cloud Registration

H-V2X: A Large Scale Highway Dataset for BEV Perception

Dataset Enhancement with Instance-Level Augmentations

RealGen: Retrieval Augmented Generation for Controllable Traffic Scenarios

OmniNOCS: A unified NOCS dataset and model for 3D lifting of 2D objects

VETRA: A Dataset for Vehicle Tracking in Aerial Imagery - New Challenges for Multi-Object Tracking

Gaussian Frosting: Editable Complex Radiance Fields with Real-Time Rendering

Omni-Recon: Harnessing Image-based Rendering for General-Purpose Neural Radiance Fields

Smoothness, Synthesis, and Sampling: Re-thinking Unsupervised Multi-View Stereo with DIV Loss

FisherRF: Active View Selection and Mapping with Radiance Fields using Fisher Information

Learning 3D-aware GANs from Unposed Images with Template Feature Field

Analytic-Splatting: Anti-Aliased 3D Gaussian Splatting via Analytic Integration

AttentionHand: Text-driven Controllable Hand Image Generation for 3D Hand Reconstruction in the Wild

MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images

Generative Camera Dolly: Extreme Monocular Dynamic Novel View Synthesis

ConDense: Consistent 2D-3D Pre-training for Dense and Sparse Features from Multi-View Images

Animal Avatars: Reconstructing Animatable 3D Animals from Casual Videos

Rasterized Edge Gradients: Handling Discontinuities Differentially

LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation

Pyramid Diffusion for Fine 3D Large Scene Generation

MobileNetV4: Universal Models for the Mobile Ecosystem

Adaptive Parametric Activation

Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models

Decoupling Common and Unique Representations for Multimodal Self-supervised Learning

GiT: Towards Generalist Vision Transformer through Universal Language Interface

Mamba-ND: Selective State Space Modeling for Multi-Dimensional Data

LongVLM: Efficient Long Video Understanding via Large Language Models

A Simple Low-bit Quantization Framework for Video Snapshot Compressive Imaging

Minimalist Vision with Freeform Pixels

POET: Prompt Offset Tuning for Continual Human Action Adaptation

An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models

Denoising Vision Transformers

PiTe: Pixel-Temporal Alignment for Large Video-Language Model

Sapiens: Foundation for Human Vision Models

Turbo: Informativity-Driven Acceleration Plug-In for Vision-Language Large Models

Improving Knowledge Distillation via Regularizing Feature Direction and Norm

Chameleon: A Data-Efficient Generalist for Dense Visual Prediction in the Wild