Representing Long Volumetric Video with Temporal Gaussian Hierarchy

This paper aims to address the challenge of reconstructing long volumetric videos from multi-view RGB videos. Recent dynamic view synthesis methods leverage powerful 4D representations, like feature grids or point cloud sequences, to achieve high-quality rendering results. However, they are typically limited to short (1~2s) video clips and often suffer from large memory footprints when dealing with longer videos. To solve this issue, we propose a novel 4D representation, named Temporal Gaussian Hierarchy, to compactly model long volumetric videos. Our key observation is that there are generally various degrees of temporal redundancy in dynamic scenes, which consist of areas changing at different speeds. Motivated by this, our approach builds a multi-level hierarchy of 4D Gaussian primitives, where each level separately describes scene regions with different degrees of content change, and adaptively shares Gaussian primitives to represent unchanged scene content over different temporal segments, thus effectively reducing the number of Gaussian primitives. In addition, the tree-like structure of the Gaussian hierarchy allows us to efficiently represent the scene at a particular moment with a subset of Gaussian primitives, leading to nearly constant GPU memory usage during the training or rendering regardless of the video length. Extensive experimental results demonstrate the superiority of our method over alternative methods in terms of training cost, rendering speed, and storage usage. To our knowledge, this work is the first approach capable of efficiently handling minutes of volumetric video data while maintaining state-of-the-art rendering quality. Our project page is available at: https://zju3dv.github.io/longvolcap.

World-Grounded Human Motion Recovery via Gravity-View Coordinates

We present a novel method for recovering world-grounded human motion from monocular video. The main challenge lies in the ambiguity of defining the world coordinate system, which varies between sequences. Previous approaches attempt to alleviate this issue by predicting relative motion in an autoregressive manner, but are prone to accumulating errors. Instead, we propose estimating human poses in a novel Gravity-View (GV) coordinate system, which is defined by the world gravity and the camera view direction. The proposed GV system is naturally gravity-aligned and uniquely defined for each video frame, largely reducing the ambiguity of learning image-pose mapping. The estimated poses can be transformed back to the world coordinate system using camera rotations, forming a global motion sequence. Additionally, the per-frame estimation avoids error accumulation in the autoregressive methods. Experiments on in-the-wild benchmarks demonstrate that our method recovers more realistic motion in both the camera space and world-grounded settings, outperforming state-of-the-art methods in both accuracy and speed. The code is available at https://zju3dv.github.io/gvhmr/.

MSFMamba: Multi-Scale Feature Fusion State Space Model for Multi-Source Remote Sensing Image Classification

In multi-source remote sensing image classification field, remarkable progress has been made by convolutional neural network and Transformer. However, existing methods are still limited due to the inherent local reductive bias. Recently, Mamba-based methods built upon the State Space Model have shown great potential for long-range dependency modeling with linear complexity, but it has rarely been explored for the multi-source remote sensing image classification task. To this end, we propose Multi-Scale Feature Fusion Mamba (MSFMamba) network for hyperspectral image (HSI) and LiDAR/SAR data joint classification. Specifically, MSFMamba mainly comprises three parts: Multi-Scale Spatial Mamba (MSpa-Mamba) block, Spectral Mamba (Spe-Mamba) block, and Fusion Mamba (Fus-Mamba) block. Specifically, to solve the feature redundancy in multiple canning routes, the MSpa-Mamba block incorporates the multi-scale strategy to minimize the computational redundancy and alleviate the feature redundancy of SSM. In addition, Spe-Mamba is designed for spectral feature exploration, which is essential for HSI feature modeling. Moreover, to alleviate the heterogeneous gap between HSI and LiDAR/SAR data, we design Fus-Mamba block for multi-source feature fusion. The original Mamba is extended to accommodate dual inputs, and cross-modal feature interaction is enhanced. Extensive experimental results on three multi-source remote sensing datasets demonstrate the superiority performance of the proposed MSFMamba over the state-of-the-art models. Source codes of MSFMamba will be made public available at https://github.com/summitgao/MSFMamba .

Wavelet-based Bi-dimensional Aggregation Network for SAR Image Change Detection

Synthetic aperture radar (SAR) image change detection is critical in remote sensing image analysis. Recently, the attention mechanism has been widely used in change detection tasks. However, existing attention mechanisms often employ down-sampling operations such as average pooling on the Key and Value components to enhance computational efficiency. These irreversible operations result in the loss of high-frequency components and other important information. To address this limitation, we develop Wavelet-based Bi-dimensional Aggregation Network (WBANet) for SAR image change detection. We design a wavelet-based self-attention block that includes discrete wavelet transform and inverse discrete wavelet transform operations on Key and Value components. Hence, the feature undergoes downsampling without any loss of information, while simultaneously enhancing local contextual awareness through an expanded receptive field. Additionally, we have incorporated a bi-dimensional aggregation module that boosts the non-linear representation capability by merging spatial and channel information via broadcast mechanism. Experimental results on three SAR datasets demonstrate that our WBANet significantly outperforms contemporary state-of-the-art methods. Specifically, our WBANet achieves 98.33\%, 96.65\%, and 96.62\% of percentage of correct classification (PCC) on the respective datasets, highlighting its superior performance. Source codes are available at \url{https://github.com/summitgao/WBANet}.

LSKSANet: A Novel Architecture for Remote Sensing Image Semantic Segmentation Leveraging Large Selective Kernel and Sparse Attention Mechanism

In this paper, we proposed large selective kernel and sparse attention network (LSKSANet) for remote sensing image semantic segmentation. The LSKSANet is a lightweight network that effectively combines convolution with sparse attention mechanisms. Specifically, we design large selective kernel module to decomposing the large kernel into a series of depth-wise convolutions with progressively increasing dilation rates, thereby expanding the receptive field without significantly increasing the computational burden. In addition, we introduce the sparse attention to keep the most useful self-attention values for better feature aggregation. Experimental results on the Vaihingen and Postdam datasets demonstrate the superior performance of the proposed LSKSANet over state-of-the-art methods.

Generating Human Motion in 3D Scenes from Text Descriptions

Generating human motions from textual descriptions has gained growing research interest due to its wide range of applications. However, only a few works consider human-scene interactions together with text conditions, which is crucial for visual and physical realism. This paper focuses on the task of generating human motions in 3D indoor scenes given text descriptions of the human-scene interactions. This task presents challenges due to the multi-modality nature of text, scene, and motion, as well as the need for spatial reasoning. To address these challenges, we propose a new approach that decomposes the complex problem into two more manageable sub-problems: (1) language grounding of the target object and (2) object-centric motion generation. For language grounding of the target object, we leverage the power of large language models. For motion generation, we design an object-centric scene representation for the generative model to focus on the target object, thereby reducing the scene complexity and facilitating the modeling of the relationship between human motions and the object. Experiments demonstrate the better motion quality of our approach compared to baselines and validate our design choices.

MaPa: Text-driven Photorealistic Material Painting for 3D Shapes

This paper aims to generate materials for 3D meshes from text descriptions. Unlike existing methods that synthesize texture maps, we propose to generate segment-wise procedural material graphs as the appearance representation, which supports high-quality rendering and provides substantial flexibility in editing. Instead of relying on extensive paired data, i.e., 3D meshes with material graphs and corresponding text descriptions, to train a material graph generative model, we propose to leverage the pre-trained 2D diffusion model as a bridge to connect the text and material graphs. Specifically, our approach decomposes a shape into a set of segments and designs a segment-controlled diffusion model to synthesize 2D images that are aligned with mesh parts. Based on generated images, we initialize parameters of material graphs and fine-tune them through the differentiable rendering module to produce materials in accordance with the textual description. Extensive experiments demonstrate the superior performance of our framework in photorealism, resolution, and editability over existing methods. Project page: https://zhanghe3z.github.io/MaPa/

IntrinsicAnything: Learning Diffusion Priors for Inverse Rendering Under Unknown Illumination

This paper aims to recover object materials from posed images captured under an unknown static lighting condition. Recent methods solve this task by optimizing material parameters through differentiable physically based rendering. However, due to the coupling between object geometry, materials, and environment lighting, there is inherent ambiguity during the inverse rendering process, preventing previous methods from obtaining accurate results. To overcome this ill-posed problem, our key idea is to learn the material prior with a generative model for regularizing the optimization process. We observe that the general rendering equation can be split into diffuse and specular shading terms, and thus formulate the material prior as diffusion models of albedo and specular. Thanks to this design, our model can be trained using the existing abundant 3D object data, and naturally acts as a versatile tool to resolve the ambiguity when recovering material representations from RGB images. In addition, we develop a coarse-to-fine training strategy that leverages estimated materials to guide diffusion models to satisfy multi-view consistent constraints, leading to more stable and accurate results. Extensive experiments on real-world and synthetic datasets demonstrate that our approach achieves state-of-the-art performance on material recovery. The code will be available at https://zju3dv.github.io/IntrinsicAnything.

SpatialTracker: Tracking Any 2D Pixels in 3D Space

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.

Hybrid Convolutional and Attention Network for Hyperspectral Image Denoising

Hyperspectral image (HSI) denoising is critical for the effective analysis and interpretation of hyperspectral data. However, simultaneously modeling global and local features is rarely explored to enhance HSI denoising. In this letter, we propose a hybrid convolution and attention network (HCANet), which leverages both the strengths of convolution neural networks (CNNs) and Transformers. To enhance the modeling of both global and local features, we have devised a convolution and attention fusion module aimed at capturing long-range dependencies and neighborhood spectral correlations. Furthermore, to improve multi-scale information aggregation, we design a multi-scale feed-forward network to enhance denoising performance by extracting features at different scales. Experimental results on mainstream HSI datasets demonstrate the rationality and effectiveness of the proposed HCANet. The proposed model is effective in removing various types of complex noise. Our codes are available at \url{https://github.com/summitgao/HCANet}.