Skip to content
Free planNo credit card requiredStart writing
AI for Research home
AI for Research home
Dashboard
Sign inGet started

Search 250 million research papers

Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.

About 10,000 papers for “multi-scale representation”

Multi-Scale Representation Learning on Proteins

Vignesh Ram Somnath, Charlotte Bunne, Andreas Krause · arXiv (Cornell University) · 2022

Proteins are fundamental biological entities mediating key roles in cellular function and disease. This paper introduces a multi-scale graph construction of a protein -- HoloProt -- connecting surface to structure and sequence. The surface captures coarser details of the protein, while sequence as primary component and structure -- comprising secondary and tertiary components -- capture finer details. Our graph encoder then learns a multi-scale representation by allowing each level to integrate the encoding from level(s) below with the graph at that level. We test the learned representation on different tasks, (i.) ligand binding affinity (regression), and (ii.) protein function prediction (classification). On the regression task, contrary to previous methods, our model performs consistently and reliably across different dataset splits, outperforming all baselines on most splits. On the classification task, it achieves a performance close to the top-performing model while using 10x fewer parameters. To improve the memory efficiency of our construction, we segment the multiplex protein surface manifold into molecular superpixels and substitute the surface with these superpixels at little to no performance loss.

Cited by 22Open access

Multi-Scale Representation Learning on Proteins

Vignesh Ram Somnath, Charlotte Bunne, Andreas Krause · Zenodo (CERN European Organization for Nuclear Research) · 2021

Protein-ligand structures from PDBBind (v2019), and protein structures from the enzyme dataset as described in the paper: "Multi-Scale Representation Learning on Proteins" with associated code at https://github.com/vsomnath/holoprot. <strong>Paper Abstract</strong> Proteins are fundamental biological entities mediating key roles in cellular function and disease. This paper introduces a multi-scale graph construction of a protein – HoloProt – connecting surface to structure and sequence. The surface captures coarser details of the protein, while sequence as primary component and structure – comprising secondary and tertiary components – capture finer details. Our graph encoder then learns a multi-scale representation by allowing each level to integrate the encoding from level(s) below with the graph at that level. We test the learned representation on different tasks, (i.) ligand binding affinity (regression), and (ii.) protein function prediction (classification). On the regression task, contrary to previous methods, our model performs consistently and reliably across different dataset splits, outperforming all baselines on most splits. On the classification task, it achieves a performance close to the top-performing model while using 10x fewer parameters. To improve the memory efficiency of our construction, we segment the multiplex protein surface manifold into molecular superpixels and substitute the surface with these superpixels at little to no performance loss.

Cited by 12Open access

Multi-Scale Fusion for Object Representation

Rongzhen Zhao, Vivienne Wang, Juho Kannala et al. · arXiv (Cornell University) · 2024

Representing images or videos as object-level feature vectors, rather than pixel-level feature maps, facilitates advanced visual tasks. Object-Centric Learning (OCL) primarily achieves this by reconstructing the input under the guidance of Variational Autoencoder (VAE) intermediate representation to drive so-called \textit{slots} to aggregate as much object information as possible. However, existing VAE guidance does not explicitly address that objects can vary in pixel sizes while models typically excel at specific pattern scales. We propose \textit{Multi-Scale Fusion} (MSF) to enhance VAE guidance for OCL training. To ensure objects of all sizes fall within VAE's comfort zone, we adopt the \textit{image pyramid}, which produces intermediate representations at multiple scales; To foster scale-invariance/variance in object super-pixels, we devise \textit{inter}/\textit{intra-scale fusion}, which augments low-quality object super-pixels of one scale with corresponding high-quality super-pixels from another scale. On standard OCL benchmarks, our technique improves mainstream methods, including state-of-the-art diffusion-based ones. The source code is available on https://github.com/Genera1Z/MultiScaleFusion.

Cited by 0Open access

Multi-Scale Trees Representation for Object’s Skeletons

Xiao Chen · 2004

Object representation based on skeleton is of importance in computer vision. Now many skeletonization methods based on different theory appear, however the studies on efficiently describing and recognizing objects with skeletons are limited. An algorithm constructing the hierarchical tree representation based on skeleton elements by top-down decomposition of object skeletons is presented in this paper. By introducing the conception of scale, the multi-scale trees with fewer nodes and stable concatenations are acquired. Experiments show that the tree representation is compact and robust and able of lowering the complexity of graph matching.

Cited by 0

Knowledge-guided Land Use Multi-scale Representation

Junbo Yu, Daniel Xiao, Wenbin Wang · Abstracts of the ICA · 2025

Remotely sensed imagery and field surveying have substantially contributed to the universal availability of land-use maps.Presently, multi-scale cartography techniques have also facilitated visual representations of land use.However, the multitemporal, multi-scale, and multi-category characteristics of these data often lead to diversity and complexity in crossscale outcomes (Zhu et al., 2022).The main factors are due to: (1) the single format of traditional fundamental geomatics products and the lag in updating at the map sectors; (2) the fragmented abstract representation and modeling at separate scales; and (3) the complicated intertwined geometric patterns, topological correlations and semantic attributes of landuse data at different scales.These cartographic mission discrepancies and incomplete perception regarding target scales may contribute to misjudgments in macro decision-making, overall planning, and mapping exports on land information.In Figure 1, for example, the results are misleading land information visualizations at different scales in some inappropriate ways.The centerline extraction at the end of the braided bayous is not satisfactory when smoothing the lake boundaries.New redundant regions may be triggered at adjacent land use necks during scaling.

Cited by 0Open access

Multi-Scale Representation of a Digital Line

Benjamin Zhan, Barbara P. Buttenfield · Cartography and Geographic Information Systems · 1996

A scheme for multi-scale geometrical representation of a digital line is proposed. A digital line is represented as a structured object in which lines at different levels of resolution are represented as lists, quadcodes are used as control mechanism for guiding line simplification at multiple scales of a pyramid, and operators for simplification and smoothing are used to generate lines at different levels of resolution. In this scheme, lines at different levels of an image pyramid can be automatically generated without specifying the so-called tolerance value. Experimental results show that up to a certain level of resolution, the simplified lines retain important information content and recognizable characteristics. Also, the image pyramid assures a legible representation at any level of representation. Future research should improve the current line simplification operator and address the correspondence between lines at different levels of an image pyramid and the lines at different map scales.

Cited by 12

Multi-scale 3D representation via volumetric quasi-random scale space

Akshaya Mishra, Alexander Wong, Paul Fieguth et al. · 2011

A novel nonlinear volumetric scale-space framework is proposed for multi-scale volumetric data representation. The problem is formulated as a Bayesian least-squares estimator, and a quasi-random density estimation approach is introduced for estimating the posterior distribution between consecutive volumetric scale space realizations. Experimental results using both synthetic and real MR volumetric data demonstrate the effectiveness of the proposed scale-space framework for three-dimensional representation with significantly better structural separation and localization across all scales when compared to existing volumetric scale-space frameworks such as volumetric anisotropic diffusion and volumetric linear Gaussian scale-space, especially under scenarios with high noise levels.

Cited by 0

Robust Multi-Stage Tracking via Multi-Scale and Multi-Level Representation Learning

Ning Li, Bineng Zhong, Qihua Liang et al. · IEEE Transactions on Multimedia · 2025

How to learn multi-scale and multi-level representations is crucial for robust tracking. However, most current one-stream structure based trackers with visual transformers (dubbed ViTs) cannot effectively capture multi-scale representations due to the structure of their adopted ViTs is non-hierarchical. Meanwhile, they often only use the output features from the final layer for predicting results (i.e., ignoring the utilization of low-level features from the shallow layers) which may result in a certain degree of lacking multi-level representation learning ability. To address these issues, we propose a robust multi-stage tracker that effectively combines the advantages of both hierarchical and one-stream structured ViT as a tracking backbone to improve the multi-scale and multi-level representation learning abilities. Specifically, first of all, we design a hierarchical tracker with a three-stage backbone. In the first two stages of our tracker, we utilize a dual-branch structure to obtain multi-scale features of the template and search region separately. Especially, We design the local scale awareness modules based on simple MLP layers to capture multi-scale features. These modules remove complex operations such as convolutions or shifted window attentions, thus avoiding the performance degradation caused by traditional hierarchical ViTs. In the third stage (i.e. the main stage), we construct a global encoder based on the one-stream ViT to achieve efficient feature extraction and feature interaction for our tracker. Then, we design a multi-level feature integration module in the main stage to explicitly utilize the representation information learned from the shallow layers and fuse them with the features of the final layer to obtain multi-level representation information. Lastly, benefit from the these designs, our tracker can effectively capture more multi-scale and multi-level representations for robust tracking. Comprehensive experiments on GOT-10k, LaSOT, LaSOT<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$_{ext}$</tex-math></inline-formula>, TNL2K, UAV123, TrackingNet and VOT2020 benchmarks validate the effectiveness and robustness of our method.

Cited by 0

Multi-Scale Deep Representation Aggregation for Vein Recognition

Zaiyu Pan, Jun Wang, Guoqing Wang et al. · IEEE Transactions on Information Forensics and Security · 2020

The recent success of Deep Convolutional Neural Network (DCNN) for various computer vision tasks such as image recognition has already demonstrated its robust feature representation ability. However, the limitation of training database on small scale vein recognition tasks restricts its performance because the recognition result of DCNN depends heavily on the number of trainsets. This motivates the design of a Multi-Scale Deep Representation Aggregation (MSDRA) model based on a pre-trained DCNN for vein recognition. First, the multi-scale feature maps are extracted by a pre-trained DCNN model. Second, a local mean threshold approach is designed to preliminarily remove the noisy information of multi-scale feature maps and generate the selected feature maps. Third, we propose an Unsupervised Vein Information Mining (UVIM) method to localize vein information of selected feature maps for generating a binary vein information mask, and then the vein information mask is utilized to keep useful deep representation and discard the background information. Finally, the discriminative multi-scale deep representations, which are generated by using the vein information mask to aggregate multi-scale feature maps, are concatenated into the final compact feature vectors, and then a Support Vector Machine (SVM) is introduced for final recognition. Our proposed model outperforms the state-of-the-art methods on two benchmark vein databases. Moreover, an additional experiment using the subset of PolyU Palmprint database illustrates the system's generalization ability and robustness.

Cited by 22

Multi-scale Sparse Representation for Robust Face Recognition

Mao X. Nguyen, Quang Minh Le, Vu Hong Son Pham et al. · 2011

Recently the Sparse Representation-based Classification (SRC) has been successfully used in face recognition. In SRC, a test image is coded by a linear combination of the training dictionary. In this paper, we propose a model extends from SRC named Multi-scale SRC (MSRC). The MSRC build the multi-scale dictionary for the training. A test image is then coded using this multi-scale dictionary. In addition, a voting scheme is applied which not only helps improving the recognition rate significantly, but also makes the algorithm more robust with occlusion. Experiments on representative face databases demonstrate that the MSRC is much more effective than the SRC.

Cited by 16

Implicit Neural Representation with Multi-Scale Sine Activation

Jufeng Han, Shu Wei, Min Wu et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 2026

Implicit Neural Representations (INRs) have become a powerful paradigm for modeling continuous signals in computer vision, graphics, and scientific computing. However, multilayer perceptrons (MLPs) generally suffer from severe spectral bias, which limits their ability to accurately model high-frequency details and multi-scale structures. To address this challenge, we propose a novel Multi-Scale Sine Activation (MSA), which explicitly introduces multi-scale frequency responses by incorporating multiple sets of sine activations with logarithmically spaced frequencies in parallel at each layer. MSA is further combined with an amplitude modulation mechanism to ensure numerical stability and robust optimization across different frequency channels. We conduct extensive experiments on a series of challenging tasks, including 1D multi-scale function fitting, image representation, video representation, 3D shape representation, and PDEs solving. Experimental results show that MSA outperforms existing state-of-the-art methods in terms of reconstruction accuracy, detail preservation, and training stability.

Cited by 0Open access

Schematic representation of the multi-scale liver architecture.

Luciana Melina Luque (14725510), Carlos Manuel Carlevaro (14725513), Camilo Julio Llamoza Torres (14725516) et al. · Figshare · 2023

&lt;p&gt;The human liver is divided grossly into four parts or lobes. The four lobes are the right lobe, the left lobe, the caudate lobe, and the quadrate lobe. Seen from the front the liver is divided into two lobes: the right lobe and the left lobe. It is further divided in eight functionally independent segments based in the the Couinaud classification of liver anatomy. At the microscopic (histological) scale, the liver is organized in repetitive functional units called liver lobules, which take the shape of polygonal prisms (typically hexagonal in cross section). Each lobule is mainly constituted by hepatocytes and it is centered on a branch of the hepatic vein called the central vein which is interconnected with the interlobular portal triads: the hepatic artery (red), the portal vein (blue), bile duct (green).&lt;/p&gt;

Cited by 0Open access

Multi-Scale Deep Representation Learning in Synthetic Biology

Liu, Xiaoyi · Scholar Commons (University of South Carolina) · 2024

Synthetic biology advances and combines the expertise of engineers and biologists, bridging the gap between engineering and natural life. Synthetic biology has been generally categorized into two broad branches by developing new biological components, networks, and systems to reprogram organisms. The first branch involves using synthetic molecules to mimic natural biological functions. The second branch focuses on assembling natural biological components in novel ways, aiming to produce systems with unique, practical functions. Thus, the de novo engineering of biological modules and synthetic pathways is used in related practical bioengineering applications, such as drug-targeting strategies and microbial product manufacturing. Therefore, synthetic biology represents a new paradigm in scientific exploration and innovation, with widely used implications for our understanding and optimization of biological systems. Over the past decades, there has been a significant increase in the amount of available whole-genome sequencing data and experimental data due to the emergence of new automation technologies, such as high-content imaging, high-throughput screening, and sequencing. Given the growth of these data sets, researchers are unable to summarize these data simply from experience and memory. Thus, stable and efficient computational methods are required to integrate them to predict or reveal new phenomena or insights that have never been discovered. However, incomplete knowledge of metabolic processes impairs the accuracy of biological systems, hindering advancements in systems biology and metabolic engineering. Additionally, some fundamental challenges still remain. Firstly, problems in systems biology are often cross-scale and multi-modal, yet existing computational methods for problem definition and model design are often single-scale and single-modal. Secondly, biological systems are multi-scale, unbalanced, and noisy, making structuring and benchmarking this complicated data very difficult. Thirdly, most natural or valuable products' complete biosynthetic pathways are unknown. Thus, computer-aided biosynthesis planning holds significant value. To address the above challenges, we introduce multi-scale deep learning-based representation learning methodologies to understand and optimize the downstream tasks in systems biology, such as metabolic pathway inference, missing reaction prediction in GEMs, and retrosynthesis prediction. Specifically, our first study introduces a novel Multi-View Multi-Label learning framework for Metabolic Pathway Inference (MVML-MPI), which outperforms State-Of-The-Art (SOTA) methods by accurately representing the complex relationships between compounds and pathways. In the second study, to address the limitation of incomplete metabolic knowledge in GEMs, we proposed a novel framework named hypergraph Convolution network and attention mechanism integrated Explorer for GAPS prediction of metabolism termed CLOSEgaps. It is a comprehensive deep learning-driven tool that represents the hyper-topological information of GEMs and effectively fills gaps through hyperlink prediction, thereby enhancing the accuracy of phenotypic predictions. In the third study, we proposed a novel end-to-end framework for one-step retrosynthesis that combines the power of a graph encoder, which integrates learnable structural information, with the capability to sequentially translate drugs, thereby efficiently capturing chemically plausible information (RetroCaptioner). This research presents an advancement in systems biology by introducing a suite of multi-scale deep learning methodologies. These methodologies tackle key challenges such as MVML-MPI enhancing our understanding of complex metabolic pathways, CLOSEgaps innovatively filling gaps in metabolic models, and RetroCaptioner facilitating the process of retrosynthesis. Taken together, they form a comprehensive and integrated approach, and our proposed methods significantly advance the capabilities of synthetic biology.

Cited by 0