Decoupling Bidirectional Geometric Representations of 4D cost volume with 2D convolution
High-performance real-time stereo matching methods invariably rely on 3D regularization of the cost volume, which is unfriendly to mobile devices. And 2D regularization based methods struggle in ill-posed regions. In this paper, we present a deployment-friendly 4D cost aggregation network DBStereo, which is based on pure 2D convolutions. Specifically, we first provide a thorough analysis of the decoupling characteristics of 4D cost volume. And design a lightweight bidirectional geometry aggregation block to capture spatial and disparity representation respectively. Through decoupled learning, our approach achieves real-time performance and impressive accuracy simultaneously. Extensive experiments demonstrate that our proposed DBStereo outperforms all existing aggregation-based methods in both inference time and accuracy, even surpassing the iterative-based method IGEV-Stereo. Our study break the empirical design of using 3D convolutions for 4D cost volume and provides a simple yet strong baseline of the proposed decouple aggregation paradigm for further study. Code will be available at (\href{https://github.com/happydummy/DBStereo}{https://github.com/happydummy/DBStereo}) soon.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Unleashing Semantic and Geometric Priors for 3D Scene Completion
Camera-based 3D semantic scene completion (SSC) provides dense geometric and semantic perception for autonomous driving and robotic navigation. However, existing methods rely on a coupled encoder to deliver both semantic…
3D Semantic Scene CompletionAutonomous DrivingBack to the Future: Bidirectional Information Decoupling Network for Multi-turn Dialogue Modeling
Multi-turn dialogue modeling as a challenging branch of natural language understanding (NLU), aims to build representations for machines to understand human dialogues, which provides a solid foundation for multiple downs…
Natural Language UnderstandingBiGS: Bidirectional Gaussian Primitives for Relightable 3D Gaussian Splatting
We present Bidirectional Gaussian Primitives, an image-based novel view synthesis technique designed to represent and render 3D objects with surface and volumetric materials under dynamic illumination. Our approach integ…
Novel View SynthesisSAND: Spatially Adaptive Network Depth for Fast Sampling of Neural Implicit Surfaces
Implicit neural representations are powerful for geometric modeling, but their practical use is often limited by the high computational cost of network evaluations. We observe that implicit representations require progre…
MonoJSG: Joint Semantic and Geometric Cost Volume for Monocular 3D Object Detection
Due to the inherent ill-posed nature of 2D-3D projection, monocular 3D object detection lacks accurate depth recovery ability. Although the deep neural network (DNN) enables monocular depth-sensing from high-level learne…
3D Object DetectionDepth EstimationMonocular 3D Object DetectionObject+2