Photo-realistic free-viewpoint rendering of real-world scenes using classical computer graphics techniques is challenging, because it requires the difficult step of capturing detailed appearance and geometry models. Recent studies have demonstrated promising results by learning scene representations that implicitly encode both geometry and appearance without 3D supervision. However, existing approaches in practice often show blurry renderings caused by the limited network capacity or the difficulty in finding accurate intersections of camera rays with the scene geometry. Synthesizing high-resolution imagery from these representations often requires time-consuming optical ray marching. In this work, we introduce Neural Sparse Voxel Fields (NSVF), a new neural scene representation for fast and high-quality free-viewpoint rendering. NSVF defines a set of voxel-bounded implicit fields organized in a sparse voxel octree to model local properties in each cell. We progressively learn the underlying voxel structures with a differentiable ray-marching operation from only a set of posed RGB images. With the sparse voxel octree structure, rendering novel views can be accelerated by skipping the voxels containing no relevant scene content. Our method is typically over 10 times faster than the state-of-the-art (namely, NeRF(Mildenhall et al., 2020)) at inference time while achieving higher quality results. Furthermore, by utilizing an explicit sparse voxel representation, our method can easily be applied to scene editing and scene composition. We also demonstrate several challenging tasks, including multi-scene learning, free-viewpoint rendering of a moving human, and large-scale scene rendering. Code and data are available at our website: https://github.com/facebookresearch/NSVF.
GeoSVR: Taming Sparse Voxels for Geometrically Accurate Surface Reconstruction
School of Computer Science and Engineering, State Key Laboratory of Complex Critical Software Environment, Jiangxi Research Institute, Beihang University（北京航空航天大学计算机科学与工程学院、复杂临界软件环境国家重点实验室、江西研究院）；Rawmantic AI；State Key Laboratory of Virtual Reality Technology and Systems, Beijing（虚拟现实技术与系统国家重点实验室）；Macquarie University（麦考瑞大学）；RIKEN AIP（日本理化学研究所AIP）；The University of Tokyo（东京大学）
Surfsvr: 2D Surface Priors as 3D Geometric Regularizers for Sparse Voxel Reconstruction
HIT (Shenzhen)（哈尔滨工业大学（深圳））；Tsinghua University（清华大学）；NWPU（西北工业大学）；Beijing Institute of Control and Engineering（北京控制与工程研究所）；Technical University of Munich（慕尼黑工业大学）
NYU（纽约大学）；NVIDIA（英伟达）；ASU（亚利桑那州立大学）；University of Toronto（多伦多大学）；Vector Institute（向量研究所）；Caltech（加州理工学院）
Humans can easily imagine the complete 3D geometry of occluded objects and scenes. This appealing ability is vital for recognition and understanding. To enable such capability in AI systems, we propose VoxFormer, a Transformer-based semantic scene completion framework that can output complete 3D volumetric semantics from only 2D images. Our framework adopts a two-stage design where we start from a sparse set of visible and occupied voxel queries from depth estimation, followed by a densification stage that generates dense 3D voxels from the sparse ones. A key idea of this design is that the vi…
Peking University（北京大学）；Max Planck Institute for Informatics（马克斯·普朗克信息学研究所）；Huawei（华为）；Zhejiang Lab（之江实验室）
Designing an efficient yet deployment-friendly 3D backbone to handle sparse point clouds is a fundamental problem in 3D perception. Compared with the customized sparse convolution, the attention mechanism in Transformers is more appropriate for flexibly modeling long-range relationships and is easier to be deployed in real-world applications. However, due to the sparse characteristics of point clouds, it is non-trivial to apply a standard transformer on sparse points. In this paper, we present Dynamic Sparse Voxel Transformer (DSVT), a single-stride window-based voxel Transformer backbone for …
Nanjing University of Science and Technology（南京理工大学）
3D object tracking in point clouds is still a challenging problem due to the sparsity of LiDAR points in dynamic environments. In this work, we propose a Siamese voxel-to-BEV tracker, which can significantly improve the tracking performance in sparse 3D point clouds. Specifically, it consists of a Siamese shape-aware feature learning network and a voxel-to-BEV target localization network. The Siamese shape-aware feature learning network can capture 3D shape information of the object to learn the discriminative features of the object so that the potential target from the background in sparse po…
Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）；School of Computer Science and Engineering, Nanjing University of Science and Technology（南京理工大学计算机科学与工程学院）；Department of Management Science and Engineering, Tsinghua University（清华大学管理科学与工程系）
Point cloud-based large scale place recognition is fundamental for many applications like Simultaneous Localization and Mapping (SLAM). Although many models have been proposed and have achieved good performance by learning short-range local features, long-range contextual properties have often been neglected. Moreover, the model size has also become a bottleneck for their wide applications. To overcome these challenges, we propose a super light-weight network model termed SVT-Net for large scale place recognition. Specifically, on top of the highly efficient 3D Sparse Convolution (SP-Conv), an…
Hong Kong University of Science and Technology（香港科技大学）
Recently, many approaches have been proposed through single or multiple representations to improve the performance of point cloud semantic segmentation. However, these works do not maintain a good balance among performance, efficiency, and memory consumption. To address these issues, we propose DRINet++ that extends DRINet by enhancing the sparsity and geometric properties of a point cloud with a voxel-as-point principle. To improve efficiency and performance, DRINet++ mainly consists of two modules: Sparse Feature Encoder and Sparse Geometry Feature Enhancement. The Sparse Feature Encoder ext…
Huazhong University of Science & Technology（华中科技大学）
Rendering moving human bodies at free viewpoints only from a monocular video is quite a challenging problem. The information is too sparse to model complicated human body structures and motions from both view and pose dimensions. Neural radiance fields (NeRF) have shown great power in novel view synthesis and have been applied to human body rendering. However, most current NeRF-based methods bear huge costs for both training and rendering, which impedes the wide applications in real-life scenarios. In this paper, we propose a rendering framework that can learn moving human body structures extr…
SPARF: Large-Scale Learning of 3D Sparse Radiance Fields from Few Input Images
Technical University of Munich (TUM)（慕尼黑工业大学（TUM））；King Abdullah University of Science and Technology (KAUST)（阿卜杜拉国王科技大学（KAUST））；University of Oxford（牛津大学）
HUST（华中科技大学）；Huawei Inc.（华为公司）；Technical University of Munich（慕尼黑工业大学）
Neural radiance fields (NeRF) have shown great success in modeling 3D scenes and synthesizing novel-view images. However, most previous NeRF methods take much time to optimize one single scene. Explicit data structures, e.g. voxel features, show great potential to accelerate the training process. However, voxel features face two big challenges to be applied to dynamic scenes, i.e. modeling temporal information and capturing different scales of point motions. We propose a radiance field framework by representing scenes with time-aware voxel features, named as TiNeuVox. A tiny coordinate deforma…
LiDAR is an important method for autonomous driving systems to sense the environment. The point clouds obtained by LiDAR typically exhibit sparse and irregular distribution, thus posing great challenges to the detection of 3D objects, especially those that are small and distant. To tackle this difficulty, we propose Reconfigurable Voxels, a new approach to constructing representations from 3D point clouds. Specifically, we devise a biased random walk scheme, which adaptively covers each neighborhood with a fixed number of voxels based on the local spatial distribution and produces a representa…
TiAVox: Time-aware Attenuation Voxels for Sparse-view 4D DSA Reconstruction
Huazhong University of Science and Technology（华中科技大学）；Union Hospital, Tongji Medical College, Huazhong University of Science and Technology（华中科技大学同济医学院附属协和医院）
Nanjing University（南京大学）；HangZhou Normal University（杭州师范大学）；University of Missouri（密苏里大学）；Jiangsu Longyuan Zhenhua Marine Engineering Co., LTD.（江苏龙源振华海洋工程有限公司）
This study develops a unified Point Cloud Geometry (PCG) compression method through the processing of multiscale sparse tensor-based voxelized PCG. We call this compression method SparsePCGC. The proposed SparsePCGC is a low complexity solution because it only performs the convolutions on sparsely-distributed Most-Probable Positively-Occupied Voxels (MP-POV). The multiscale representation also allows us to compress scale-wise MP-POVs by exploiting cross-scale and same-scale correlations extensively and flexibly. The overall compression efficiency highly depends on the accuracy of estimated occ…
University of Birmingham（伯明翰大学）；Toshiba Research Europe（东芝欧洲研究院）
For intelligent robotics applications, extending 3D mapping to 3D semantic mapping enables robots to, not only localize themselves with respect to the scene's geometrical features but also simultaneously understand the higher level meaning of the scene contexts. Most previous methods focus on geometric 3D reconstruction and scene understanding independently notwithstanding the fact that joint estimation can boost the accuracy of the semantic mapping. In this paper, a dense RGB-D semantic mapping system with a Pixel-Voxel network is proposed, which can perform dense 3D mapping while simultaneou…
Indian Institute of Information Technology Pune（印度信息技术学院浦那分院）
Hand pose estimation from monocular depth images has been an important and challenging problem in the Computer Vision community. In this paper, we present a novel approach to estimate 3D hand joint locations from 2D depth images. Unlike most of the previous methods, our model captures the 3D spatial information from a depth image thereby giving it a greater understanding of the input. We voxelize the input depth map to capture the 3D features of the input and perform 3D data augmentations to make our network robust to real-world images. Our network is trained in an end-to-end manner which redu…
Deep generative models have shown success in generating 3D shapes with different representations. In this work, we propose Neural Volumetric Mesh Generator(NVMG) which can generate novel and high-quality volumetric meshes. Unlike the previous 3D generative model for point cloud, voxel, and implicit surface, the volumetric mesh representation is a ready-to-use representation in industry with details on both the surface and interior. Generating this such highly-structured data thus brings a significant challenge. We first propose a diffusion-based generative model to tackle this problem by gener…
This paper presents a learning-based, lossless compression method for static point cloud geometry, based on context-adaptive arithmetic coding. Unlike most existing methods working in the octree domain, our encoder operates in a hybrid mode, mixing octree and voxel-based coding. We adaptively partition the point cloud into multi-resolution voxel blocks according to the point cloud structure, and use octree to signal the partitioning. On the one hand, octree representation can eliminate the sparsity in the point cloud. On the other hand, in the voxel domain, convolutions can be naturally expres…
Hangzhou Dianzi University（杭州电子科技大学）；University College London（伦敦大学学院）
The recently developed pure Transformer architectures have attained promising accuracy on point cloud learning benchmarks compared to convolutional neural networks. However, existing point cloud Transformers are computationally expensive since they waste a significant amount of time on structuring the irregular data. To solve this shortcoming, we present Sparse Window Attention (SWA) module to gather coarse-grained local features from non-empty voxels, which not only bypasses the expensive irregular data structuring and invalid empty voxel computation, but also obtains linear computational com…
VoxNeRF: Bridging Voxel Representation and Neural Radiance Fields for Enhanced Indoor View Synthesis
Technical University of Munich（慕尼黑工业大学）；Huawei Munich Research Center（华为慕尼黑研究中心）；Munich Center for Machine Learning（慕尼黑机器学习中心）；University of Stuttgart（斯图加特大学）；Karlsruhe University of Applied Sciences（卡尔斯鲁厄应用技术大学）
NeuroVoxel-LM: Language-Aligned 3D Perception via Dynamic Voxelization and Meta-Embedding
School of Electrical and Electronic Engineering（电气与电子工程学院）；Nanyang Technological University（南洋理工大学）；School of Computer Science and Technology（计算机科学与技术学院）；University of Chinese Academy of Sciences（中国科学院大学）
Hikvision Research Institute（海康威视研究院）；Zhejiang University（浙江大学）
Point clouds can be represented in many forms (views), typically, point-based sets, voxel-based cells or range-based images(i.e., panoramic view). The point-based view is geometrically accurate, but it is disordered, which makes it difficult to find local neighbors efficiently. The voxel-based view is regular, but sparse, and computation grows cubically when voxel resolution increases. The range-based view is regular and generally dense, however spherical projection makes physical dimensions distorted. Both voxel- and range-based views suffer from quantization loss, especially for voxels when …
This paper describes a novel lossless point cloud compression algorithm that uses a neural network for estimating the coding probabilities for the occupancy status of voxels, depending on wide three dimensional contexts around the voxel to be encoded. The point cloud is represented as an octree, with each resolution layer being sequentially encoded and decoded using arithmetic coding, starting from the lowest resolution, until the final resolution is reached. The occupancy probability of each voxel of the splitting pattern at each node of the octree is modeled by a neural network, having at it…
Recently, Neural Radiance Fields (NeRF) is revolutionizing the task of novel view synthesis (NVS) for its superior performance. In this paper, we propose to synthesize dynamic scenes. Extending the methods for static scenes to dynamic scenes is not straightforward as both the scene geometry and appearance change over time, especially under monocular setup. Also, the existing dynamic NeRF methods generally require a lengthy per-scene training procedure, where multi-layer perceptrons (MLP) are fitted to model both motions and radiance. In this paper, built on top of the recent advances in voxel-…
SCube: Instant Large-Scale Scene Reconstruction using VoxSplats
NVIDIA（英伟达）；University of Toronto（多伦多大学）；Shanghai Jiao Tong University（上海交通大学）；University of Cambridge（剑桥大学）；National University of Singapore（新加坡国立大学）；Vector Institute（矢量研究所）
University of Tsukuba（筑波大学）；University of Tokyo（东京大学）；Google Brain（谷歌大脑）；University of Toronto（多伦多大学）
Neural Radiance Field (NeRF) is a popular method in data-driven 3D reconstruction. Given its simplicity and high quality rendering, many NeRF applications are being developed. However, NeRF's big limitation is its slow speed. Many attempts are made to speeding up NeRF training and inference, including intricate code-level optimization and caching, use of sophisticated data structures, and amortization through multi-task and meta learning. In this work, we revisit the basic building blocks of NeRF through the lens of classic techniques before NeRF. We propose Voxel-Accelearated NeRF (VaxNeRF), …
Mines ParisTech（巴黎高等矿业学院）；PSL Research University（巴黎文理研究大学）；Centre for Robotics（机器人中心）
In this article we describe a new convolutional neural network (CNN) to classify 3D point clouds of urban or indoor scenes. Solutions are given to the problems encountered working on scene point clouds, and a network is described that allows for point classification using only the position of points in a multi-scale neighborhood. On the reduced-8 Semantic3D benchmark [Hackel et al., 2017], this network, ranked second, beats the state of the art of point classification methods (those not using a regularization step).
The Chinese University of Hong Kong（香港中文大学）；The University of Hong Kong（香港大学）；MEGVII Technology（旷视科技）
3D object detectors usually rely on hand-crafted proxies, e.g., anchors or centers, and translate well-studied 2D frameworks to 3D. Thus, sparse voxel features need to be densified and processed by dense prediction heads, which inevitably costs extra computation. In this paper, we instead propose VoxelNext for fully sparse 3D object detection. Our core insight is to predict objects directly based on sparse voxel features, without relying on hand-crafted proxies. Our strong sparse convolutional network VoxelNeXt detects and tracks 3D objects through voxel features entirely. It is an elegant and…
University of Electronic Science and Technology of China（电子科技大学）
Accurate 3D object detection from point clouds has become a crucial component in autonomous driving. However, the volumetric representations and the projection methods in previous works fail to establish the relationships between the local point sets. In this paper, we propose Sparse Voxel-Graph Attention Network (SVGA-Net), a novel end-to-end trainable network which mainly contains voxel-graph module and sparse-to-dense regression module to achieve comparable 3D detection tasks from raw LIDAR data. Specifically, SVGA-Net constructs the local complete graph within each divided 3D spherical vox…
Tsinghua University（清华大学）；Microsoft Research Asia（微软亚洲研究院）
Sparse voxel-based 3D convolutional neural networks (CNNs) are widely used for various 3D vision tasks. Sparse voxel-based 3D CNNs create sparse non-empty voxels from the 3D input and perform 3D convolution operations on them only. We propose a simple yet effective padding scheme --- interpolation-aware padding to pad a few empty voxels adjacent to the non-empty voxels and involve them in the 3D CNN computation so that all neighboring voxels exist when computing point-wise features via the trilinear interpolation. For fine-grained 3D vision tasks where point-wise features are essential, like s…
