Qianqian Wang

Zhengqi Li, Qianqian Wang, Forrester Cole et al. · deepmind

We address the problem of synthesizing novel views from a monocular video depicting a complex dynamic scene. State-of-the-art methods based on temporally varying Neural Radiance Fields (aka dynamic NeRFs) have shown impressive results on this task. However, for long videos with complex object motions and uncontrolled camera trajectories, these methods can produce blurry or inaccurate renderings, hampering their use in real-world applications. Instead of encoding the entire dynamic scene within the weights of MLPs, we present a new approach that addresses these limitations by adopting a volumetric image-based rendering framework that synthesizes new viewpoints by aggregating features from nearby views in a scene-motion-aware manner. Our system retains the advantages of prior methods in its ability to model complex scenes and view-dependent effects, but also enables synthesizing photo-realistic novel views from long videos featuring complex scene dynamics with unconstrained camera trajectories. We demonstrate significant improvements over state-of-the-art methods on dynamic scene datasets, and also apply our approach to in-the-wild videos with challenging camera and object motion, where prior methods fail to produce high-quality renderings. Our project webpage is at dynibar.github.io.

Qianqian Wang, Yen-Yu Chang, Ruojin Cai et al. · deepmind

We present a new test-time optimization method for estimating dense and long-range motion from a video sequence. Prior optical flow or particle video tracking algorithms typically operate within limited temporal windows, struggling to track through occlusions and maintain global consistency of estimated motion trajectories. We propose a complete and globally consistent motion representation, dubbed OmniMotion, that allows for accurate, full-length motion estimation of every pixel in a video. OmniMotion represents a video using a quasi-3D canonical volume and performs pixel-wise tracking via bijections between local and canonical space. This representation allows us to ensure global consistency, track through occlusions, and model any combination of camera and object motion. Extensive evaluations on the TAP-Vid benchmark and real-world footage show that our approach outperforms prior state-of-the-art methods by a large margin both quantitatively and qualitatively. See our project page for more results: http://omnimotion.github.io/

Haoyu Guo, Sida Peng, Haotong Lin et al.

This paper addresses the challenge of reconstructing 3D indoor scenes from multi-view images. Many previous works have shown impressive reconstruction results on textured objects, but they still have difficulty in handling low-textured planar regions, which are common in indoor scenes. An approach to solving this issue is to incorporate planer constraints into the depth map estimation in multi-view stereo-based methods, but the per-view plane estimation and depth optimization lack both efficiency and multi-view consistency. In this work, we show that the planar constraints can be conveniently integrated into the recent implicit neural representation-based reconstruction methods. Specifically, we use an MLP network to represent the signed distance function as the scene geometry. Based on the Manhattan-world assumption, planar constraints are employed to regularize the geometry in floor and wall regions predicted by a 2D semantic segmentation network. To resolve the inaccurate segmentation, we encode the semantics of 3D points with another MLP and design a novel loss that jointly optimizes the scene geometry and semantics in 3D space. Experiments on ScanNet and 7-Scenes datasets show that the proposed method outperforms previous methods by a large margin on 3D reconstruction quality. The code is available at https://zju3dv.github.io/manhattan_sdf.

Jiaming Sun, Xi Chen, Qianqian Wang et al. · deepmind

We are witnessing an explosion of neural implicit representations in computer vision and graphics. Their applicability has recently expanded beyond tasks such as shape generation and image-based rendering to the fundamental problem of image-based 3D reconstruction. However, existing methods typically assume constrained 3D environments with constant illumination captured by a small set of roughly uniformly distributed cameras. We introduce a new method that enables efficient and accurate surface reconstruction from Internet photo collections in the presence of varying illumination. To achieve this, we propose a hybrid voxel- and surface-guided sampling technique that allows for more efficient ray sampling around surfaces and leads to significant improvements in reconstruction quality. Further, we present a new benchmark and protocol for evaluating reconstruction performance on such in-the-wild scenes. We perform extensive experiments, demonstrating that our approach surpasses both classical and neural reconstruction methods on a wide variety of metrics.

Luming Tang, Menglin Jia, Qianqian Wang et al. · deepmind

Finding correspondences between images is a fundamental problem in computer vision. In this paper, we show that correspondence emerges in image diffusion models without any explicit supervision. We propose a simple strategy to extract this implicit knowledge out of diffusion networks as image features, namely DIffusion FeaTures (DIFT), and use them to establish correspondences between real images. Without any additional fine-tuning or supervision on the task-specific data or annotations, DIFT is able to outperform both weakly-supervised methods and competitive off-the-shelf features in identifying semantic, geometric, and temporal correspondences. Particularly for semantic correspondence, DIFT from Stable Diffusion is able to outperform DINO and OpenCLIP by 19 and 14 accuracy points respectively on the challenging SPair-71k benchmark. It even outperforms the state-of-the-art supervised methods on 9 out of 18 categories while remaining on par for the overall performance. Project page: https://diffusionfeatures.github.io

Zhengqi Li, Qianqian Wang, Noah Snavely et al. · deepmind

We present a method for learning to generate unbounded flythrough videos of natural scenes starting from a single view, where this capability is learned from a collection of single photographs, without requiring camera poses or even multiple views of each scene. To achieve this, we propose a novel self-supervised view generation training paradigm, where we sample and rendering virtual camera trajectories, including cyclic ones, allowing our model to learn stable view generation from a collection of single views. At test time, despite never seeing a video during training, our approach can take a single image and generate long camera trajectories comprised of hundreds of new views with realistic and diverse content. We compare our approach with recent state-of-the-art supervised view generation methods that require posed multi-view videos and demonstrate superior performance and synthesis quality.

Sida Peng, Zhen Xu, Junting Dong et al.

This paper addresses the challenge of reconstructing an animatable human model from a multi-view video. Some recent works have proposed to decompose a non-rigidly deforming scene into a canonical neural radiance field and a set of deformation fields that map observation-space points to the canonical space, thereby enabling them to learn the dynamic scene from images. However, they represent the deformation field as translational vector field or SE(3) field, which makes the optimization highly under-constrained. Moreover, these representations cannot be explicitly controlled by input motions. Instead, we introduce a pose-driven deformation field based on the linear blend skinning algorithm, which combines the blend weight field and the 3D human skeleton to produce observation-to-canonical correspondences. Since 3D human skeletons are more observable, they can regularize the learning of the deformation field. Moreover, the pose-driven deformation field can be controlled by input skeletal motions to generate new deformation fields to animate the canonical human model. Experiments show that our approach significantly outperforms recent human modeling methods. The code is available at https://zju3dv.github.io/animatable_nerf/.

Ruojin Cai, Joseph Tung, Qianqian Wang et al. · deepmind

We consider the visual disambiguation task of determining whether a pair of visually similar images depict the same or distinct 3D surfaces (e.g., the same or opposite sides of a symmetric building). Illusory image matches, where two images observe distinct but visually similar 3D surfaces, can be challenging for humans to differentiate, and can also lead 3D reconstruction algorithms to produce erroneous results. We propose a learning-based approach to visual disambiguation, formulating it as a binary classification task on image pairs. To that end, we introduce a new dataset for this problem, Doppelgangers, which includes image pairs of similar structures with ground truth labels. We also design a network architecture that takes the spatial distribution of local keypoints and matches as input, allowing for better reasoning about both local and global cues. Our evaluation shows that our method can distinguish illusory matches in difficult cases, and can be integrated into SfM pipelines to produce correct, disambiguated 3D reconstructions. See our project page for our code, datasets, and more results: http://doppelgangers-3d.github.io/.

Qianqian Wang, Zhengqi Li, David Salesin et al. · deepmind

We introduce 3D Moments, a new computational photography effect. As input we take a pair of near-duplicate photos, i.e., photos of moving subjects from similar viewpoints, common in people's photo collections. As output, we produce a video that smoothly interpolates the scene motion from the first photo to the second, while also producing camera motion with parallax that gives a heightened sense of 3D. To achieve this effect, we represent the scene as a pair of feature-based layered depth images augmented with scene flow. This representation enables motion interpolation along with independent control of the camera viewpoint. Our system produces photorealistic space-time videos with motion parallax and scene dynamics, while plausibly recovering regions occluded in the original views. We conduct extensive experiments demonstrating superior performance over baselines on public datasets and in-the-wild photos. Project page: https://3d-moments.github.io/

Haotong Lin, Qianqian Wang, Ruojin Cai et al. · deepmind

In this work, we aim to reconstruct a time-varying 3D model, capable of rendering photo-realistic renderings with independent control of viewpoint, illumination, and time, from Internet photos of large-scale landmarks. The core challenges are twofold. First, different types of temporal changes, such as illumination and changes to the underlying scene itself (such as replacing one graffiti artwork with another) are entangled together in the imagery. Second, scene-level temporal changes are often discrete and sporadic over time, rather than continuous. To tackle these problems, we propose a new scene representation equipped with a novel temporal step function encoding method that can model discrete scene-level content changes as piece-wise constant functions over time. Specifically, we represent the scene as a space-time radiance field with a per-image illumination embedding, where temporally-varying scene changes are encoded using a set of learned step functions. To facilitate our task of chronology reconstruction from Internet imagery, we also collect a new dataset of four scenes that exhibit various changes over time. We demonstrate that our method exhibits state-of-the-art view synthesis results on this dataset, while achieving independent control of viewpoint, time, and illumination.

Qianqian Wang, Vickie Ye, Hang Gao et al.

Monocular dynamic reconstruction is a challenging and long-standing vision problem due to the highly ill-posed nature of the task. Existing approaches depend on templates, are effective only in quasi-static scenes, or fail to model 3D motion explicitly. We introduce a method for reconstructing generic dynamic scenes, featuring explicit, persistent 3D motion trajectories in the world coordinate frame, from casually captured monocular videos. We tackle the problem with two key insights: First, we exploit the low-dimensional structure of 3D motion by representing scene motion with a compact set of SE(3) motion bases. Each point's motion is expressed as a linear combination of these bases, facilitating soft decomposition of the scene into multiple rigidly-moving groups. Second, we take advantage of off-the-shelf data-driven priors such as monocular depth maps and long-range 2D tracks, and devise a method to effectively consolidate these noisy supervisory signals, resulting in a globally consistent representation of the dynamic scene. Experiments show that our method achieves state-of-the-art performance for both long-range 3D/2D motion estimation and novel view synthesis on dynamic scenes. Project Page: https://shape-of-motion.github.io/

Justin Kerr, Chung Min Kim, Mingxuan Wu et al.

Humans can learn to manipulate new objects by simply watching others; providing robots with the ability to learn from such demonstrations would enable a natural interface specifying new behaviors. This work develops Robot See Robot Do (RSRD), a method for imitating articulated object manipulation from a single monocular RGB human demonstration given a single static multi-view object scan. We first propose 4D Differentiable Part Models (4D-DPM), a method for recovering 3D part motion from a monocular video with differentiable rendering. This analysis-by-synthesis approach uses part-centric feature fields in an iterative optimization which enables the use of geometric regularizers to recover 3D motions from only a single video. Given this 4D reconstruction, the robot replicates object trajectories by planning bimanual arm motions that induce the demonstrated object part motion. By representing demonstrations as part-centric trajectories, RSRD focuses on replicating the demonstration's intended behavior while considering the robot's own morphological limits, rather than attempting to reproduce the hand's motion. We evaluate 4D-DPM's 3D tracking accuracy on ground truth annotated 3D part trajectories and RSRD's physical execution performance on 9 objects across 10 trials each on a bimanual YuMi robot. Each phase of RSRD achieves an average of 87% success rate, for a total end-to-end success rate of 60% across 90 trials. Notably, this is accomplished using only feature fields distilled from large pretrained vision models -- without any task-specific training, fine-tuning, dataset collection, or annotation. Project page: https://robot-see-robot-do.github.io

Qianqian Wang, Wei Wang, Yuqi Fang et al.

Resting-state functional magnetic resonance imaging (rs-fMRI) can reflect spontaneous neural activities in brain and is widely used for brain disorder analysis.Previous studies propose to extract fMRI representations through diverse machine/deep learning methods for subsequent analysis. But the learned features typically lack biological interpretability, which limits their clinical utility. From the view of graph theory, the brain exhibits a remarkable modular structure in spontaneous brain functional networks, with each module comprised of functionally interconnected brain regions-of-interest (ROIs). However, most existing learning-based methods for fMRI analysis fail to adequately utilize such brain modularity prior. In this paper, we propose a Brain Modularity-constrained dynamic Representation learning (BMR) framework for interpretable fMRI analysis, consisting of three major components: (1) dynamic graph construction, (2) dynamic graph learning via a novel modularity-constrained graph neural network(MGNN), (3) prediction and biomarker detection for interpretable fMRI analysis. Especially, three core neurocognitive modules (i.e., salience network, central executive network, and default mode network) are explicitly incorporated into the MGNN, encouraging the nodes/ROIs within the same module to share similar representations. To further enhance discriminative ability of learned features, we also encourage the MGNN to preserve the network topology of input graphs via a graph topology reconstruction constraint. Experimental results on 534 subjects with rs-fMRI scans from two datasets validate the effectiveness of the proposed method. The identified discriminative brain ROIs and functional connectivities can be regarded as potential fMRI biomarkers to aid in clinical diagnosis.

Jing Li, Quanxue Gao, Qianqian Wang et al.

Multi-view clustering (MVC) based on non-negative matrix factorization (NMF) and its variants have received a huge amount of attention in recent years due to their advantages in clustering interpretability. However, existing NMF-based multi-view clustering methods perform NMF on each view data respectively and ignore the impact of between-view. Thus, they can't well exploit the within-view spatial structure and between-view complementary information. To resolve this issue, we present semi-non-negative tensor factorization (Semi-NTF) and develop a novel multi-view clustering based on Semi-NTF with one-side orthogonal constraint. Our model directly performs Semi-NTF on the 3rd-order tensor which is composed of anchor graphs of views. Thus, our model directly considers the between-view relationship. Moreover, we use the tensor Schatten p-norm regularization as a rank approximation of the 3rd-order tensor which characterizes the cluster structure of multi-view data and exploits the between-view complementary information. In addition, we provide an optimization algorithm for the proposed method and prove mathematically that the algorithm always converges to the stationary KKT point. Extensive experiments on various benchmark datasets indicate that our proposed method is able to achieve satisfactory clustering performance.

Yuqi Fang, Jinjian Wu, Qianqian Wang et al.

Resting-state functional MRI (rs-fMRI) is increasingly employed in multi-site research to aid neurological disorder analysis. Existing studies usually suffer from significant cross-site/domain data heterogeneity caused by site effects such as differences in scanners/protocols. Many methods have been proposed to reduce fMRI heterogeneity between source and target domains, heavily relying on the availability of source data. But acquiring source data is challenging due to privacy concerns and/or data storage burdens in multi-site studies. To this end, we design a source-free collaborative domain adaptation (SCDA) framework for fMRI analysis, where only a pretrained source model and unlabeled target data are accessible. Specifically, a multi-perspective feature enrichment method (MFE) is developed for target fMRI analysis, consisting of multiple collaborative branches to dynamically capture fMRI features of unlabeled target data from multiple views. Each branch has a data-feeding module, a spatiotemporal feature encoder, and a class predictor. A mutual-consistency constraint is designed to encourage pair-wise consistency of latent features of the same input generated from these branches for robust representation learning. To facilitate efficient cross-domain knowledge transfer without source data, we initialize MFE using parameters of a pretrained source model. We also introduce an unsupervised pretraining strategy using 3,806 unlabeled fMRIs from three large-scale auxiliary databases, aiming to obtain a general feature encoder. Experimental results on three public datasets and one private dataset demonstrate the efficacy of our method in cross-scanner and cross-study prediction tasks. The model pretrained on large-scale rs-fMRI data has been released to the public.

Junhao Zhang, Qianqian Wang, Xiaochuan Wang et al.

Resting-state functional magnetic resonance imaging (rs-fMRI) offers a non-invasive approach to examining abnormal brain connectivity associated with brain disorders. Graph neural network (GNN) gains popularity in fMRI representation learning and brain disorder analysis with powerful graph representation capabilities. Training a general GNN often necessitates a large-scale dataset from multiple imaging centers/sites, but centralizing multi-site data generally faces inherent challenges related to data privacy, security, and storage burden. Federated Learning (FL) enables collaborative model training without centralized multi-site fMRI data. Unfortunately, previous FL approaches for fMRI analysis often ignore site-specificity, including demographic factors such as age, gender, and education level. To this end, we propose a specificity-aware federated graph learning (SFGL) framework for rs-fMRI analysis and automated brain disorder identification, with a server and multiple clients/sites for federated model aggregation and prediction. At each client, our model consists of a shared and a personalized branch, where parameters of the shared branch are sent to the server while those of the personalized branch remain local. This can facilitate knowledge sharing among sites and also helps preserve site specificity. In the shared branch, we employ a spatio-temporal attention graph isomorphism network to learn dynamic fMRI representations. In the personalized branch, we integrate vectorized demographic information (i.e., age, gender, and education years) and functional connectivity networks to preserve site-specific characteristics. Representations generated by the two branches are then fused for classification. Experimental results on two fMRI datasets with a total of 1,218 subjects suggest that SFGL outperforms several state-of-the-art approaches.

Mengyu Xu, Qiaoxin Yang, Qianqian Wang et al.

Large language models (LLMs) are increasingly used to provide public-facing health information, yet existing safety evaluations overlook whether responses preserve comparable medical information across different user phrasings of the same question. To address this, we introduce the Medical Information Response Audit (MIRA), a bilingual, controlled benchmark that assesses whether LLMs provide comparable medical information across user-side language, register, and health literacy signals. MIRA contains 4,320 prompts built from 60 medically reviewed, low-risk health questions. Across five mainstream LLMs, models answered all medical questions, but responses to low health-literacy signals consistently omitted more key information, provided fewer concrete next steps, and offered less support for independent judgment. We term this pattern Differential Information Dilution (DID). Language effects are model-specific rather than uniformly worse for non-English prompts. A comparison with 300 real-world health queries provides preliminary evidence of rank-order validity. A knowledge-guided mitigation prompt reduces information dilution for most models, with the largest reductions in underinformative simplification observed for Claude (~8%) and Qwen (~6%).

Qitao Zhao, Hao Tan, Qianqian Wang et al.

Self-supervised pre-training has revolutionized foundation models for languages, individual 2D images and videos, but remains largely unexplored for learning 3D-aware representations from multi-view images. In this paper, we present E-RayZer, a self-supervised large 3D Vision model that learns truly 3D-aware representations directly from unlabeled images. Unlike prior self-supervised methods such as RayZer that infer 3D indirectly through latent-space view synthesis, E-RayZer operates directly in 3D space, performing self-supervised 3D reconstruction with Explicit geometry. This formulation eliminates shortcut solutions and yields representations that are geometrically grounded. To ensure convergence and scalability, we introduce a novel fine-grained learning curriculum that organizes training from easy to hard samples and harmonizes heterogeneous data sources in an entirely unsupervised manner. Experiments demonstrate that E-RayZer significantly outperforms RayZer on pose estimation, matches or sometimes surpasses fully supervised reconstruction models such as VGGT. Furthermore, its learned representations outperform leading visual pre-training models (e.g., DINOv3, CroCo v2, VideoMAE V2, and RayZer) when transferring to 3D downstream tasks, establishing E-RayZer as a new paradigm for 3D-aware visual pre-training.

Haiwen Feng, Junyi Zhang, Qianqian Wang et al.

Dynamic 3D reconstruction and point tracking in videos are typically treated as separate tasks, despite their deep connection. We propose St4RTrack, a feed-forward framework that simultaneously reconstructs and tracks dynamic video content in a world coordinate frame from RGB inputs. This is achieved by predicting two appropriately defined pointmaps for a pair of frames captured at different moments. Specifically, we predict both pointmaps at the same moment, in the same world, capturing both static and dynamic scene geometry while maintaining 3D correspondences. Chaining these predictions through the video sequence with respect to a reference frame naturally computes long-range correspondences, effectively combining 3D reconstruction with 3D tracking. Unlike prior methods that rely heavily on 4D ground truth supervision, we employ a novel adaptation scheme based on a reprojection loss. We establish a new extensive benchmark for world-frame reconstruction and tracking, demonstrating the effectiveness and efficiency of our unified, data-driven framework. Our code, model, and benchmark will be released.

Boyang Ding, Chenglong Chu, Dunju Zang et al.

Autoregressive large language models (LLMs) scale well by expressing diverse tasks as sequences of discrete natural-language tokens and training with next-token prediction, which unifies comprehension and generation under self-supervision. Extending this paradigm to multimodal data requires a shared, discrete representation across modalities. However, most vision-language models (VLMs) still rely on a hybrid interface: discrete text tokens paired with continuous Vision Transformer (ViT) features. Because supervision is largely text-driven, these models are often biased toward understanding and cannot fully leverage large-scale self-supervised learning on non-text data. Recent work has explored discrete visual tokenization to enable fully autoregressive multimodal modeling, showing promising progress toward unified understanding and generation. Yet existing discrete vision tokens frequently lose information due to limited code capacity, resulting in noticeably weaker understanding than continuous-feature VLMs. We present Kelix, a fully discrete autoregressive unified model that closes the understanding gap between discrete and continuous visual representations.

Aditya Chetan, Eric Cai, Peeyush Kushwaha et al.

The emergence of Large Vision-Language Models (LVLMs) has significantly advanced video understanding capabilities. However, existing benchmarks focus predominantly on coarse-grained tasks such as action segmentation, classification, captioning, and retrieval. Furthermore, these benchmarks often rely on entities that can be easily identified verbally, like household objects, animals, human subjects, etc., limiting their applicability to complex, in-the-wild video scenarios. But, many applications such as furniture assembly, cooking, etc., require step-by-step fine-grained spatio-temporal understanding of the video, which is not sufficiently evaluated in current benchmarks. To address this gap, we introduce Flat-Pack Bench, a novel benchmark centered on furniture assembly tasks. Our benchmark evaluates LVLMs on nuanced tasks, including temporal ordering of assembly actions, temporal localization of assembly state, understanding part mating, and tracking, using multiple-choice questions paired with visual prompts highlighting relevant parts as references for fine-grained questions. Our experiments reveal that state-of-the-art LVLMs struggle significantly with fine-grained spatio-temporal reasoning, highlighting their limitations in effectively leveraging temporal information from videos, limited tracking ability, and understanding of spatial interactions like physical contact.

Nan Huang, Wenzhao Zheng, Chenfeng Xu et al.

Moving object segmentation is a crucial task for achieving a high-level understanding of visual scenes and has numerous downstream applications. Humans can effortlessly segment moving objects in videos. Previous work has largely relied on optical flow to provide motion cues; however, this approach often results in imperfect predictions due to challenges such as partial motion, complex deformations, motion blur and background distractions. We propose a novel approach for moving object segmentation that combines long-range trajectory motion cues with DINO-based semantic features and leverages SAM2 for pixel-level mask densification through an iterative prompting strategy. Our model employs Spatio-Temporal Trajectory Attention and Motion-Semantic Decoupled Embedding to prioritize motion while integrating semantic support. Extensive testing on diverse datasets demonstrates state-of-the-art performance, excelling in challenging scenarios and fine-grained segmentation of multiple objects. Our code is available at https://motion-seg.github.io/.

Yuqi Fang, Junhao Zhang, Linmin Wang et al.

Functional magnetic resonance imaging (fMRI) has been increasingly employed to investigate functional brain activity. Many fMRI-related software/toolboxes have been developed, providing specialized algorithms for fMRI analysis. However, existing toolboxes seldom consider fMRI data augmentation, which is quite useful, especially in studies with limited or imbalanced data. Moreover, current studies usually focus on analyzing fMRI using conventional machine learning models that rely on human-engineered fMRI features, without investigating deep learning models that can automatically learn data-driven fMRI representations. In this work, we develop an open-source toolbox, called Augmentation and Computation Toolbox for braIn netwOrk aNalysis (ACTION), offering comprehensive functions to streamline fMRI analysis. The ACTION is a Python-based and cross-platform toolbox with graphical user-friendly interfaces. It enables automatic fMRI augmentation, covering blood-oxygen-level-dependent (BOLD) signal augmentation and brain network augmentation. Many popular methods for brain network construction and network feature extraction are included. In particular, it supports constructing deep learning models, which leverage large-scale auxiliary unlabeled data (3,800+ resting-state fMRI scans) for model pretraining to enhance model performance for downstream tasks. To facilitate multi-site fMRI studies, it is also equipped with several popular federated learning strategies. Furthermore, it enables users to design and test custom algorithms through scripting, greatly improving its utility and extensibility. We demonstrate the effectiveness and user-friendliness of ACTION on real fMRI data and present the experimental results. The software, along with its source code and manual, can be accessed online.

Jisang Han, Sunghwan Hong, Jaewoo Jung et al.

Reliable 3D reconstruction from in-the-wild image collections is often hindered by "noisy" images-irrelevant inputs with little or no view overlap with others. While traditional Structure-from-Motion pipelines handle such cases through geometric verification and outlier rejection, feed-forward 3D reconstruction models lack these explicit mechanisms, leading to degraded performance under in-the-wild conditions. In this paper, we discover that the existing feed-forward reconstruction model, e.g., VGGT, despite lacking explicit outlier-rejection mechanisms or noise-aware training, can inherently distinguish distractor images. Through an in-depth analysis under varying proportions of synthetic distractors, we identify a specific layer that naturally exhibits outlier-suppressing behavior. Further probing reveals that this layer encodes discriminative internal representations that enable an effective noise-filtering capability, which we simply leverage to perform outlier-view rejection in feed-forward 3D reconstruction without any additional fine-tuning or supervision. Extensive experiments on both controlled and in-the-wild datasets demonstrate that this implicit filtering mechanism is consistent and generalizes well across diverse scenarios.

Mengqi Wu, Yongheng Sun, Qianqian Wang et al.

Aggregating multi-site brain MRI data can enhance deep learning model training, but also introduces non-biological heterogeneity caused by site-specific variations (e.g., differences in scanner vendors, acquisition parameters, and imaging protocols) that can undermine generalizability. Recent retrospective MRI harmonization seeks to reduce such site effects by standardizing image style (e.g., intensity, contrast, noise patterns) while preserving anatomical content. However, existing methods often rely on limited paired traveling-subject data or fail to effectively disentangle style from anatomy. Furthermore, most current approaches address only single-sequence harmonization, restricting their use in real-world settings where multi-sequence MRI is routinely acquired. To this end, we introduce MMH, a unified framework for multi-site multi-sequence brain MRI harmonization that leverages biomedical semantic priors for sequence-aware style alignment. MMH operates in two stages: (1) a diffusion-based global harmonizer that maps MR images to a sequence-specific unified domain using style-agnostic gradient conditioning, and (2) a target-specific fine-tuner that adapts globally aligned images to desired target domains. A tri-planar attention BiomedCLIP encoder aggregates multi-view embeddings to characterize volumetric style information, allowing explicit disentanglement of image styles from anatomy without requiring paired data. Evaluations on 4,163 T1- and T2-weighted MRIs demonstrate MMH's superior harmonization over state-of-the-art methods in image feature clustering, voxel-level comparison, tissue segmentation, and downstream age and site classification.

Qianqian Wang, Yifei Zhang, Aleksander Holynski et al. · berkeley

We present a unified framework capable of solving a broad range of 3D tasks. Our approach features a stateful recurrent model that continuously updates its state representation with each new observation. Given a stream of images, this evolving state can be used to generate metric-scale pointmaps (per-pixel 3D points) for each new input in an online fashion. These pointmaps reside within a common coordinate system, and can be accumulated into a coherent, dense scene reconstruction that updates as new images arrive. Our model, called CUT3R (Continuous Updating Transformer for 3D Reconstruction), captures rich priors of real-world scenes: not only can it predict accurate pointmaps from image observations, but it can also infer unseen regions of the scene by probing at virtual, unobserved views. Our method is simple yet highly flexible, naturally accepting varying lengths of images that may be either video streams or unordered photo collections, containing both static and dynamic content. We evaluate our method on various 3D/4D tasks and demonstrate competitive or state-of-the-art performance in each. Project Page: https://cut3r.github.io/

Yuxi Xiao, Qianqian Wang, Shangzhan Zhang et al.

Recovering dense and long-range pixel motion in videos is a challenging problem. Part of the difficulty arises from the 3D-to-2D projection process, leading to occlusions and discontinuities in the 2D motion domain. While 2D motion can be intricate, we posit that the underlying 3D motion can often be simple and low-dimensional. In this work, we propose to estimate point trajectories in 3D space to mitigate the issues caused by image projection. Our method, named SpatialTracker, lifts 2D pixels to 3D using monocular depth estimators, represents the 3D content of each frame efficiently using a triplane representation, and performs iterative updates using a transformer to estimate 3D trajectories. Tracking in 3D allows us to leverage as-rigid-as-possible (ARAP) constraints while simultaneously learning a rigidity embedding that clusters pixels into different rigid parts. Extensive evaluation shows that our approach achieves state-of-the-art tracking performance both qualitatively and quantitatively, particularly in challenging scenarios such as out-of-plane rotation.

Zhengqi Li, Richard Tucker, Forrester Cole et al. · deepmind

We present a system that allows for accurate, fast, and robust estimation of camera parameters and depth maps from casual monocular videos of dynamic scenes. Most conventional structure from motion and monocular SLAM techniques assume input videos that feature predominantly static scenes with large amounts of parallax. Such methods tend to produce erroneous estimates in the absence of these conditions. Recent neural network-based approaches attempt to overcome these challenges; however, such methods are either computationally expensive or brittle when run on dynamic videos with uncontrolled camera motion or unknown field of view. We demonstrate the surprising effectiveness of a deep visual SLAM framework: with careful modifications to its training and inference schemes, this system can scale to real-world videos of complex dynamic scenes with unconstrained camera paths, including videos with little camera parallax. Extensive experiments on both synthetic and real videos demonstrate that our system is significantly more accurate and robust at camera pose and depth estimation when compared with prior and concurrent work, with faster or comparable running times. See interactive results on our project page: https://mega-sam.github.io/

Chenglong Chu, Guorui Zhou, Guowang Zhang et al.

Long-context ability, has become one of the most important iteration direction of next-generation Large Language Models, particularly in semantic understanding/reasoning, code agentic intelligence and recommendation system. However, the standard softmax attention exhibits quadratic time complexity with respect to sequence length. As the sequence length increases, this incurs substantial overhead in long-context settings, leading the training and inference costs of extremely long sequences deteriorate rapidly. Existing solutions mitigate this issue through two technique routings: i) Reducing the KV cache per layer, such as from the head-level compression GQA, and the embedding dimension-level compression MLA, but the KV cache remains linearly dependent on the sequence length at a 1:1 ratio. ii) Interleaving with KV Cache friendly architecture, such as local attention SWA, linear kernel GDN, but often involve trade-offs among KV Cache and long-context modeling effectiveness. Besides the two technique routings, we argue that there exists an intermediate path not well explored: {Maintaining a linear relationship between the KV cache and sequence length, but performing semantic-level compression through a specific ratio $k$}. This $O(n/k)$ path does not pursue a ``minimum KV cache'', but rather trades acceptable memory costs for complete, referential, and interpretable retention of long distant dependency. Motivated by this, we propose Kwai Summary Attention (KSA), a novel attention mechanism that reduces sequence modeling cost by compressing historical contexts into learnable summary tokens.

Suyan Dai, Gan Sun, Fazeng Li et al.

Spectral clustering has emerged as one of the most effective clustering algorithms due to its superior performance. However, most existing models are designed for centralized settings, rendering them inapplicable in modern decentralized environments. Moreover, current federated learning approaches often suffer from poor generalization performance due to reliance on unreliable pseudo-labels, and fail to capture the latent correlations amongst heterogeneous clients. To tackle these limitations, this paper proposes a novel framework named Federated Multi-Task Clustering (i.e.,FMTC), which intends to learn personalized clustering models for heterogeneous clients while collaboratively leveraging their shared underlying structure in a privacy-preserving manner. More specifically, the FMTC framework is composed of two main components: client-side personalized clustering module, which learns a parameterized mapping model to support robust out-of-sample inference, bypassing the need for unreliable pseudo-labels; and server-side tensorial correlation module, which explicitly captures the shared knowledge across all clients. This is achieved by organizing all client models into a unified tensor and applying a low-rank regularization to discover their common subspace. To solve this joint optimization problem, we derive an efficient, privacy-preserving distributed algorithm based on the Alternating Direction Method of Multipliers, which decomposes the global problem into parallel local updates on clients and an aggregation step on the server. To the end, several extensive experiments on multiple real-world datasets demonstrate that our proposed FMTC framework significantly outperforms various baseline and state-of-the-art federated clustering algorithms.

Kwai Keye Team, Biao Yang, Bin Wen et al.

While Multimodal Large Language Models (MLLMs) demonstrate remarkable capabilities on static images, they often fall short in comprehending dynamic, information-dense short-form videos, a dominant medium in today's digital landscape. To bridge this gap, we introduce \textbf{Kwai Keye-VL}, an 8-billion-parameter multimodal foundation model engineered for leading-edge performance in short-video understanding while maintaining robust general-purpose vision-language abilities. The development of Keye-VL rests on two core pillars: a massive, high-quality dataset exceeding 600 billion tokens with a strong emphasis on video, and an innovative training recipe. This recipe features a four-stage pre-training process for solid vision-language alignment, followed by a meticulous two-phase post-training process. The first post-training stage enhances foundational capabilities like instruction following, while the second phase focuses on stimulating advanced reasoning. In this second phase, a key innovation is our five-mode ``cold-start'' data mixture, which includes ``thinking'', ``non-thinking'', ``auto-think'', ``think with image'', and high-quality video data. This mixture teaches the model to decide when and how to reason. Subsequent reinforcement learning (RL) and alignment steps further enhance these reasoning capabilities and correct abnormal model behaviors, such as repetitive outputs. To validate our approach, we conduct extensive evaluations, showing that Keye-VL achieves state-of-the-art results on public video benchmarks and remains highly competitive on general image-based tasks (Figure 1). Furthermore, we develop and release the \textbf{KC-MMBench}, a new benchmark tailored for real-world short-video scenarios, where Keye-VL shows a significant advantage.

Yuxi Ren, Jie Wu, Yanzuo Lu et al.

Recent advancements in diffusion-based generative image editing have sparked a profound revolution, reshaping the landscape of image outpainting and inpainting tasks. Despite these strides, the field grapples with inherent challenges, including: i) inferior quality; ii) poor consistency; iii) insufficient instrcution adherence; iv) suboptimal generation efficiency. To address these obstacles, we present ByteEdit, an innovative feedback learning framework meticulously designed to Boost, Comply, and Accelerate Generative Image Editing tasks. ByteEdit seamlessly integrates image reward models dedicated to enhancing aesthetics and image-text alignment, while also introducing a dense, pixel-level reward model tailored to foster coherence in the output. Furthermore, we propose a pioneering adversarial and progressive feedback learning strategy to expedite the model's inference speed. Through extensive large-scale user evaluations, we demonstrate that ByteEdit surpasses leading generative image editing products, including Adobe, Canva, and MeiTu, in both generation quality and consistency. ByteEdit-Outpainting exhibits a remarkable enhancement of 388% and 135% in quality and consistency, respectively, when compared to the baseline model. Experiments also verfied that our acceleration models maintains excellent performance results in terms of quality and consistency.

Biao Yang, Bin Wen, Boyang Ding et al.

In recent years, the development of Large Language Models (LLMs) has significantly advanced, extending their capabilities to multimodal tasks through Multimodal Large Language Models (MLLMs). However, video understanding remains a challenging area due to the dynamic and information-dense nature of videos. Existing models struggle with the trade-off between spatial resolution and temporal coverage when processing video content. We present Keye-VL-1.5, which addresses fundamental challenges in video comprehension through three key innovations. First, we introduce a novel Slow-Fast video encoding strategy that dynamically allocates computational resources based on inter-frame similarity, processing key frames with significant visual changes at higher resolution (Slow pathway) while handling relatively static frames with increased temporal coverage at lower resolution (Fast pathway). Second, we implement a progressive four-stage pre-training methodology that systematically extends the model's context length from 8K to 128K tokens, enabling processing of longer videos and more complex visual content. Third, we develop a comprehensive post-training pipeline focusing on reasoning enhancement and human preference alignment, incorporating a 5-step chain-of-thought data construction process, iterative GSPO-based reinforcement learning with progressive prompt hinting for difficult cases, and alignment training. Through extensive evaluation on public benchmarks and rigorous internal human assessment, Keye-VL-1.5 demonstrates significant improvements over existing models, particularly excelling in video understanding tasks while maintaining competitive performance on general multimodal benchmarks.

Nan Huang, Pengcheng Yu, Weijia Zeng et al.

Large-scale multi-view reconstruction models have made remarkable progress, but most existing approaches still rely on fully supervised training with ground-truth 3D/4D annotations. Such annotations are expensive and particularly scarce for dynamic scenes, limiting scalability. We propose SelfEvo, a self-improving framework that continually improves pretrained multi-view reconstruction models using unlabeled videos. SelfEvo introduces a self-distillation scheme using spatiotemporal context asymmetry, enabling self-improvement for learning-based 4D perception without external annotations. We systematically study design choices that make self-improvement effective, including loss signals, forms of asymmetry, and other training strategies. Across eight benchmarks spanning diverse datasets and domains, SelfEvo consistently improves pretrained baselines and generalizes across base models (e.g. VGGT and $π^3$), with significant gains on dynamic scenes. Overall, SelfEvo achieves up to 36.5% relative improvement in video depth estimation and 20.1% in camera estimation, without using any labeled data. Project Page: https://self-evo.github.io/.

Jing Li, Quanxue Gao, Qianqian Wang et al.

Multi-view clustering method based on anchor graph has been widely concerned due to its high efficiency and effectiveness. In order to avoid post-processing, most of the existing anchor graph-based methods learn bipartite graphs with connected components. However, such methods have high requirements on parameters, and in some cases it may not be possible to obtain bipartite graphs with clear connected components. To end this, we propose a label learning method based on tensor projection (LLMTP). Specifically, we project anchor graph into the label space through an orthogonal projection matrix to obtain cluster labels directly. Considering that the spatial structure information of multi-view data may be ignored to a certain extent when projected in different views separately, we extend the matrix projection transformation to tensor projection, so that the spatial structure information between views can be fully utilized. In addition, we introduce the tensor Schatten $p$-norm regularization to make the clustering label matrices of different views as consistent as possible. Extensive experiments have proved the effectiveness of the proposed method.

Shunmei Dong, Qinglong Wang, Haiqing Wang et al.

The star tracker is generally affected by the atmospheric background light and the aerodynamic environment when working in near space, which results in missing stars or false stars. Moreover, high-speed maneuvering may cause star trailing, which reduces the accuracy of the star position. To address the challenges for starmap identification, a reverse attitude statistics based method is proposed to handle position noise, false stars, and missing stars. Conversely to existing methods which match before solving for attitude, this method introduces attitude solving into the matching process, and obtains the final match and the correct attitude simultaneously by frequency statistics. Firstly, based on stable angular distance features, the initial matching is obtained by utilizing spatial hash indexing. Then, the dual-vector attitude determination is introduced to calculate potential attitude. Finally, the star pairs are accurately matched by applying a frequency statistics filtering method. In addition, Bayesian optimization is employed to find optimal parameters under the impact of noises, which is able to enhance the algorithm performance further. In this work, the proposed method is validated in simulation, field test and on-orbit experiment. Compared with the state-of-the-art, the identification rate is improved by more than 14.3%, and the solving time is reduced by over 28.5%.

Qian Jiang, Qianqian Wang, Xin Jin et al.

Remote sensing images are becoming increasingly widespread in military, earth resource exploration. Because of the limitation of a single sensor, we can obtain high spatial resolution grayscale panchromatic (PAN) images and low spatial resolution color multispectral (MS) images. Therefore, an important issue is to obtain a color image with high spatial resolution when there is only a PAN image at the input. The existing methods improve spatial resolution using super-resolution (SR) technology and spectral recovery using colorization technology. However, the SR technique cannot improve the spectral resolution, and the colorization technique cannot improve the spatial resolution. Moreover, the pansharpening method needs two registered inputs and can not achieve SR. As a result, an integrated approach is expected. To solve the above problems, we designed a novel multi-function model (MFmamba) to realize the tasks of SR, spectral recovery, joint SR and spectral recovery through three different inputs. Firstly, MFmamba utilizes UNet++ as the backbone, and a Mamba Upsample Block (MUB) is combined with UNet++. Secondly, a Dual Pool Attention (DPA) is designed to replace the skip connection in UNet++. Finally, a Multi-scale Hybrid Cross Block (MHCB) is proposed for initial feature extraction. Many experiments show that MFmamba is competitive in evaluation metrics and visual results and performs well in the three tasks when only the input PAN image is used.

Yuzhen Gao, Qianqian Wang, Yongheng Sun et al.

Accurate identification of late-life depression (LLD) using structural brain MRI is essential for monitoring disease progression and facilitating timely intervention. However, existing learning-based approaches for LLD detection are often constrained by limited sample sizes (e.g., tens), which poses significant challenges for reliable model training and generalization. Although incorporating auxiliary datasets can expand the training set, substantial domain heterogeneity, such as differences in imaging protocols, scanner hardware, and population demographics, often undermines cross-domain transferability. To address this issue, we propose a Collaborative Domain Adaptation (CDA) framework for LLD detection using T1-weighted MRIs. The CDA leverages a Vision Transformer (ViT) to capture global anatomical context and a Convolutional Neural Network (CNN) to extract local structural features, with each branch comprising an encoder and a classifier. The CDA framework consists of three stages: (a) supervised training on labeled source data, (b) self-supervised target feature adaptation and (c) collaborative training on unlabeled target data. We first train ViT and CNN on source data, followed by self-supervised target feature adaptation by minimizing the discrepancy between classifier outputs from two branches to make the categorical boundary clearer. The collaborative training stage employs pseudo-labeled and augmented target-domain MRIs, enforcing prediction consistency under strong and weak augmentation to enhance domain robustness and generalization. Extensive experiments conducted on multi-site T1-weighted MRI data demonstrate that the CDA consistently outperforms state-of-the-art unsupervised domain adaptation methods.

Meimei Yang, Yongheng Sun, Qianqian Wang et al.

Multimodal neuroimages, such as diffusion tensor imaging (DTI) and resting-state functional MRI (fMRI), offer complementary perspectives on brain activities by capturing structural or functional interactions among brain regions. While existing studies suggest that fusing these multimodal data helps detect abnormal brain activity caused by neurocognitive decline, they are generally implemented in Euclidean space and can't effectively capture intrinsic hierarchical organization of structural/functional brain networks. This paper presents a hyperbolic kernel graph fusion (HKGF) framework for neurocognitive decline analysis with multimodal neuroimages. It consists of a multimodal graph construction module, a graph representation learning module that encodes brain graphs in hyperbolic space through a family of hyperbolic kernel graph neural networks (HKGNNs), a cross-modality coupling module that enables effective multimodal data fusion, and a hyperbolic neural network for downstream predictions. Notably, HKGNNs represent graphs in hyperbolic space to capture both local and global dependencies among brain regions while preserving the hierarchical structure of brain networks. Extensive experiments involving over 4,000 subjects with DTI and/or fMRI data suggest the superiority of HKGF over state-of-the-art methods in two neurocognitive decline prediction tasks. HKGF is a general framework for multimodal data analysis, facilitating objective quantification of structural/functional brain connectivity changes associated with neurocognitive decline.

Yunshan Li, Wenwu Gong, Qianqian Wang et al.

Recent approaches based on transform-based tensor nuclear norm (TNN) have demonstrated notable effectiveness in hyperspectral image (HSI) inpainting by leveraging low-rank structures in latent representations. Recent developments incorporate deep transforms to improve low-rank tensor representation; however, existing approaches typically restrict the transform to the spectral mode, neglecting low-rank properties along other tensor modes. In this paper, we propose a novel 3-directional deep low-rank tensor representation (3DeepRep) model, which performs deep nonlinear transforms along all three modes of the HSI tensor. To enforce low-rankness, the model minimizes the nuclear norms of mode-i frontal slices in the corresponding latent space for each direction (i=1,2,3), forming a 3-directional TNN regularization. The outputs from the three directional branches are subsequently fused via a learnable aggregation module to produce the final result. An efficient gradient-based optimization algorithm is developed to solve the model in a self-supervised manner. Extensive experiments on real-world HSI datasets demonstrate that the proposed method achieves superior inpainting performance compared to existing state-of-the-art techniques, both qualitatively and quantitatively.

Yong Liu, Wenpeng Xiao, Qianqian Wang et al.

We introduce a model named DreamLight for universal image relighting in this work, which can seamlessly composite subjects into a new background while maintaining aesthetic uniformity in terms of lighting and color tone. The background can be specified by natural images (image-based relighting) or generated from unlimited text prompts (text-based relighting). Existing studies primarily focus on image-based relighting, while with scant exploration into text-based scenarios. Some works employ intricate disentanglement pipeline designs relying on environment maps to provide relevant information, which grapples with the expensive data cost required for intrinsic decomposition and light source. Other methods take this task as an image translation problem and perform pixel-level transformation with autoencoder architecture. While these methods have achieved decent harmonization effects, they struggle to generate realistic and natural light interaction effects between the foreground and background. To alleviate these challenges, we reorganize the input data into a unified format and leverage the semantic prior provided by the pretrained diffusion model to facilitate the generation of natural results. Moreover, we propose a Position-Guided Light Adapter (PGLA) that condenses light information from different directions in the background into designed light query embeddings, and modulates the foreground with direction-biased masked attention. In addition, we present a post-processing module named Spectral Foreground Fixer (SFF) to adaptively reorganize different frequency components of subject and relighted background, which helps enhance the consistency of foreground appearance. Extensive comparisons and user study demonstrate that our DreamLight achieves remarkable relighting performance.

Qianqian Wang, Wei Wang, Yuqi Fang et al.

Brain networks/graphs derived from resting-state functional MRI (fMRI) help study underlying pathophysiology of neurocognitive disorders by measuring neuronal activities in the brain. Some studies utilize learning-based methods for brain network analysis, but typically suffer from low model generalizability caused by scarce labeled fMRI data. As a notable self-supervised strategy, graph contrastive learning helps leverage auxiliary unlabeled data. But existing methods generally arbitrarily perturb graph nodes/edges to generate augmented graphs, without considering essential topology information of brain networks. To this end, we propose a topology-aware graph augmentation (TGA) framework, comprising a pretext model to train a generalizable encoder on large-scale unlabeled fMRI cohorts and a task-specific model to perform downstream tasks on a small target dataset. In the pretext model, we design two novel topology-aware graph augmentation strategies: (1) hub-preserving node dropping that prioritizes preserving brain hub regions according to node importance, and (2) weight-dependent edge removing that focuses on keeping important functional connectivities based on edge weights. Experiments on 1, 688 fMRI scans suggest that TGA outperforms several state-of-the-art methods.

Quanxue Gao, Qianqian Wang, Han Lu et al.

Although numerous clustering algorithms have been developed, many existing methods still leverage k-means technique to detect clusters of data points. However, the performance of k-means heavily depends on the estimation of centers of clusters, which is very difficult to achieve an optimal solution. Another major drawback is that it is sensitive to noise and outlier data. In this paper, from manifold learning perspective, we rethink k-means and present a new clustering algorithm which directly detects clusters of data without mean estimation. Specifically, we construct distance matrix between data points by Butterworth filter such that distance between any two data points in the same clusters equals to a small constant, while increasing the distance between other data pairs from different clusters. To well exploit the complementary information embedded in different views, we leverage the tensor Schatten p-norm regularization on the 3rd-order tensor which consists of indicator matrices of different views. Finally, an efficient alternating algorithm is derived to optimize our model. The constructed sequence was proved to converge to the stationary KKT point. Extensive experimental results indicate the superiority of our proposed method.

Yuan Liu, Sida Peng, Lingjie Liu et al.

We present a new neural representation, called Neural Ray (NeuRay), for the novel view synthesis task. Recent works construct radiance fields from image features of input views to render novel view images, which enables the generalization to new scenes. However, due to occlusions, a 3D point may be invisible to some input views. On such a 3D point, these generalization methods will include inconsistent image features from invisible views, which interfere with the radiance field construction. To solve this problem, we predict the visibility of 3D points to input views within our NeuRay representation. This visibility enables the radiance field construction to focus on visible image features, which significantly improves its rendering quality. Meanwhile, a novel consistency loss is proposed to refine the visibility in NeuRay when finetuning on a specific scene. Experiments demonstrate that our approach achieves state-of-the-art performance on the novel view synthesis task when generalizing to unseen scenes and outperforms per-scene optimization methods after finetuning.

Sida Peng, Junting Dong, Qianqian Wang et al.

This paper addresses the challenge of reconstructing an animatable human model from a multi-view video. Some recent works have proposed to decompose a non-rigidly deforming scene into a canonical neural radiance field and a set of deformation fields that map observation-space points to the canonical space, thereby enabling them to learn the dynamic scene from images. However, they represent the deformation field as translational vector field or SE(3) field, which makes the optimization highly under-constrained. Moreover, these representations cannot be explicitly controlled by input motions. Instead, we introduce neural blend weight fields to produce the deformation fields. Based on the skeleton-driven deformation, blend weight fields are used with 3D human skeletons to generate observation-to-canonical and canonical-to-observation correspondences. Since 3D human skeletons are more observable, they can regularize the learning of deformation fields. Moreover, the learned blend weight fields can be combined with input skeletal motions to generate new deformation fields to animate the human model. Experiments show that our approach significantly outperforms recent human synthesis methods. The code and supplementary materials are available at https://zju3dv.github.io/animatable_nerf/.

Kai Zhang, Fujun Luan, Qianqian Wang et al.

We present PhySG, an end-to-end inverse rendering pipeline that includes a fully differentiable renderer and can reconstruct geometry, materials, and illumination from scratch from a set of RGB input images. Our framework represents specular BRDFs and environmental illumination using mixtures of spherical Gaussians, and represents geometry as a signed distance function parameterized as a Multi-Layer Perceptron. The use of spherical Gaussians allows us to efficiently solve for approximate light transport, and our method works on scenes with challenging non-Lambertian reflectance captured under natural, static illumination. We demonstrate, with both synthetic and real data, that our reconstructions not only enable rendering of novel viewpoints, but also physics-based appearance editing of materials and illumination.

Qianqian Wang, Zhicheng Wang, Kyle Genova et al.

We present a method that synthesizes novel views of complex scenes by interpolating a sparse set of nearby views. The core of our method is a network architecture that includes a multilayer perceptron and a ray transformer that estimates radiance and volume density at continuous 5D locations (3D spatial locations and 2D viewing directions), drawing appearance information on the fly from multiple source views. By drawing on source views at render time, our method hearkens back to classic work on image-based rendering (IBR), and allows us to render high-resolution imagery. Unlike neural scene representation work that optimizes per-scene functions for rendering, we learn a generic view interpolation function that generalizes to novel scenes. We render images using classic volume rendering, which is fully differentiable and allows us to train using only multi-view posed images as supervision. Experiments show that our method outperforms recent novel view synthesis methods that also seek to generalize to novel scenes. Further, if fine-tuned on each scene, our method is competitive with state-of-the-art single-scene neural rendering methods. Project page: https://ibrnet.github.io/

Sida Peng, Yuanqing Zhang, Yinghao Xu et al.

This paper addresses the challenge of novel view synthesis for a human performer from a very sparse set of camera views. Some recent works have shown that learning implicit neural representations of 3D scenes achieves remarkable view synthesis quality given dense input views. However, the representation learning will be ill-posed if the views are highly sparse. To solve this ill-posed problem, our key idea is to integrate observations over video frames. To this end, we propose Neural Body, a new human body representation which assumes that the learned neural representations at different frames share the same set of latent codes anchored to a deformable mesh, so that the observations across frames can be naturally integrated. The deformable mesh also provides geometric guidance for the network to learn 3D representations more efficiently. To evaluate our approach, we create a multi-view dataset named ZJU-MoCap that captures performers with complex motions. Experiments on ZJU-MoCap show that our approach outperforms prior works by a large margin in terms of novel view synthesis quality. We also demonstrate the capability of our approach to reconstruct a moving person from a monocular video on the People-Snapshot dataset. The code and dataset are available at https://zju3dv.github.io/neuralbody/.

Jin Sun, Hadar Averbuch-Elor, Qianqian Wang et al.

Predicting where people can walk in a scene is important for many tasks, including autonomous driving systems and human behavior analysis. Yet learning a computational model for this purpose is challenging due to semantic ambiguity and a lack of labeled data: current datasets only tell you where people are, not where they could be. We tackle this problem by leveraging information from existing datasets, without additional labeling. We first augment the set of valid, labeled walkable regions by propagating person observations between images, utilizing 3D information to create what we call hidden footprints. However, this augmented data is still sparse. We devise a training strategy designed for such sparse labels, combining a class-balanced classification loss with a contextual adversarial loss. Using this strategy, we demonstrate a model that learns to predict a walkability map from a single image. We evaluate our model on the Waymo and Cityscapes datasets, demonstrating superior performance compared to baselines and state-of-the-art models.

Qianqian Wang, Xiaowei Zhou, Bharath Hariharan et al.

Recent research on learned visual descriptors has shown promising improvements in correspondence estimation, a key component of many 3D vision tasks. However, existing descriptor learning frameworks typically require ground-truth correspondences between feature points for training, which are challenging to acquire at scale. In this paper we propose a novel weakly-supervised framework that can learn feature descriptors solely from relative camera poses between images. To do so, we devise both a new loss function that exploits the epipolar constraint given by camera poses, and a new model architecture that makes the whole pipeline differentiable and efficient. Because we no longer need pixel-level ground-truth correspondences, our framework opens up the possibility of training on much larger and more diverse datasets for better and unbiased descriptors. We call the resulting descriptors CAmera Pose Supervised, or CAPS, descriptors. Though trained with weak supervision, CAPS descriptors outperform even prior fully-supervised descriptors and achieve state-of-the-art performance on a variety of geometric tasks. Project Page: https://qianqianwang68.github.io/CAPS/