Synthetic Objects Research Articles

Incidental learning of predictive temporal context within cortical representations of visual shape.

Abstract Incidental learning of spatio-temporal regularities and consistencies— also termed ‘statistical learning’— may be important for discovering the causal principles governing the world. We studied statistical learning of temporal structure simultaneously at two time-scales: the presentation of synthetic visual objects (3 s) and predictive temporal context (30 s) in the order of appearance of such objects. Visual objects were complex and rotated in three dimensions about varying axes. Observers viewed fifteen (15) objects recurring many times each, intermixed with other objects that appeared only once, while whole-brain BOLD activity was recorded. Over three successive days, observers grew familiar with the recurring objects and reliably distinguished them from others. As reported elsewhere (Kakaei & Braun, 2023), representational similarity analysis (RSA) of multivariate BOLD activity revealed 124 ‘object-selective’ brain parcels with selectivity for recurring objects, located mostly in the ventral occipitotemporal cortex and the parietal cortex. Here we extend RSA to the representation of predictive temporal context, specifically “temporal communities” formed by objects that tended to follow each other. After controlling for temporal proximity, we observed 27 ‘community-sensitive’ brain parcels, in which pairwise distances between multivariate responses reflected community structure, either positively (smaller distances within than between communities) or negatively (larger distances within). Among object-selective parcels, 11 parcels were positively community-sensitive in the primary visual cortex (2 parcels), the ventral occipital, lingual, or fusiform cortex (8 parcels), and the inferior temporal cortex (1 parcel). Among non-object-selective parcels, 12 parcels were negatively community-sensitive in the superior, middle, and medial frontal cortex (6 parcels), the insula (2 parcels), the putamen (1 parcel), and in the superior temporal or parietal cortex (3 parcels). We conclude that cortical representations of object shape and of predictive temporal context are largely coextensive along the ventral occipitotemporal cortex.

Gradual change of cortical representations with growing visual expertise for synthetic shapes

Abstract Objective: Visual expertise for particular categories of objects (e.g., mushrooms, birds, flowers, minerals, and so on) is known to enhance cortical responses in parts of the ventral occipitotemporal cortex. How is such additional expertise integrated into the prior cortical representation of life-long visual experience? To address this question, we presented synthetic visual objects rotating in three dimensions and recorded multivariate BOLD responses as initially unfamiliar objects gradually became familiar. Main results: An analysis of pairwise distances between multivariate BOLD responses (“representational similarity analysis,” RSA) revealed that visual objects were linearly discriminable in large parts of the ventral occipital cortex, including the primary visual cortex, as well as in certain parts of the parietal and frontal cortex. These cortical representations were present from the start, when objects were still unfamiliar, and even though objects were shown from different sides. As shapes became familiar with repeated viewing, the distribution of responses expanded to fill more of the available space. In contrast, the distribution of responses to novel shapes (which appeared only once) contracted and shifted to the margins of the available space. Conclusion: Our results revealed cortical representations of object shape and gradual changes in these representations with learning and consolidation. The cortical representations of once-viewed shapes that remained novel diverged dramatically from repeatedly viewed shapes that became familiar. This disparity was evident in both the similarity and the diversity of multivariate BOLD responses.

NeuralTO: Neural Reconstruction and View Synthesis of Translucent Objects

Learning from multi-view images using neural implicit signed distance functions shows impressive performance on 3D Reconstruction of opaque objects. However, existing methods struggle to reconstruct accurate geometry when applied to translucent objects due to the non-negligible bias in their rendering function. To address the inaccuracies in the existing model, we have reparameterized the density function of the neural radiance field by incorporating an estimated constant extinction coefficient. This modification forms the basis of our innovative framework, which is geared towards highfidelity surface reconstruction and the novel-view synthesis of translucent objects. Our framework contains two stages. In the reconstruction stage, we introduce a novel weight function to achieve accurate surface geometry reconstruction. Following the recovery of geometry, the second phase involves learning the distinct scattering properties of the participating media to enhance rendering. A comprehensive dataset, comprising both synthetic and real translucent objects, has been built for conducting extensive experiments. Experiments reveal that our method outperforms existing approaches in terms of reconstruction and novel-view synthesis.

QDETRv: Query-Guided DETR for One-Shot Object Localization in Videos

In this work, we study one-shot video object localization problem that aims to localize instances of unseen objects in the target video using a single query image of the object. Toward addressing this challenging problem, we extend a popular and successful object detection method, namely DETR (Detection Transformer), and introduce a novel approach –query-guided detection transformer for videos (QDETRv). A distinctive feature of QDETRv is its capacity to exploit information from the query image and spatio-temporal context of the target video, which significantly aids in precisely pinpointing the desired object in the video. We incorporate cross-attention mechanisms that capture temporal relationships across adjacent frames to handle the dynamic context in videos effectively. Further, to ensure strong initialization for QDETRv, we also introduce a novel unsupervised pretraining technique tailored to videos. This involves training our model on synthetic object trajectories with an analogous objective as the query-guided localization task. During this pretraining phase, we incorporate recurrent object queries and loss functions that encourage accurate patch feature reconstruction. These additions enable better temporal understanding and robust representation learning. Our experiments show that the proposed model significantly outperforms the competitive baselines on two public benchmarks, VidOR and ImageNet-VidVRD, extended for one-shot open-set localization tasks.

A Region-Adaptive Local Perturbation-Based Method for Generating Adversarial Examples in Synthetic Aperture Radar Object Detection

In synthetic aperture radar (SAR) imaging, intelligent object detection methods are facing significant challenges in terms of model robustness and application security, which are posed by adversarial examples. The existing adversarial example generation methods for SAR object detection can be divided into two main types: global perturbation attacks and local perturbation attacks. Due to the dynamic changes and irregular spatial distribution of SAR coherent speckle backgrounds, the attack effectiveness of global perturbation attacks is significantly reduced by coherent speckle. In contrast, by focusing on the image objects, local perturbation attacks achieve targeted and effective advantages over global perturbations by minimizing interference from the SAR coherent speckle background. However, the adaptability of conventional local perturbations is limited because they employ a fixed size without considering the diverse sizes and shapes of SAR objects under various conditions. This paper presents a framework for region-adaptive local perturbations (RaLP) specifically designed for SAR object detection tasks. The framework consists of two modules. To address the issue of coherent speckle noise interference in SAR imagery, we develop a local perturbation generator (LPG) module. By filtering the original image, this module reduces the speckle features introduced during perturbation generation. It then superimposes adversarial perturbations in the form of local perturbations on areas of the object with weaker speckles, thereby reducing the mutual interference between coherent speckles and adversarial perturbation. To address the issue of insufficient adaptability in terms of the size variation in local adversarial perturbations, we propose an adaptive perturbation optimizer (APO) module. This optimizer adapts the size of the adversarial perturbations based on the size and shape of the object, effectively solving the problem of adaptive perturbation size and enhancing the universality of the attack. The experimental results show that RaLP reduces the detection accuracy of the YOLOv3 detector by 29.0%, 29.9%, and 32.3% on the SSDD, SAR-Ship, and AIR-SARShip datasets, respectively, and the model-to-model and dataset-to-dataset transferability of RaLP attacks are verified.

CCDN-DETR: A Detection Transformer Based on Constrained Contrast Denoising for Multi-Class Synthetic Aperture Radar Object Detection.

The effectiveness of the SAR object detection technique based on Convolutional Neural Networks (CNNs) has been widely proven, and it is increasingly used in the recognition of ship targets. Recently, efforts have been made to integrate transformer structures into SAR detectors to achieve improved target localization. However, existing methods rarely design the transformer itself as a detector, failing to fully leverage the long-range modeling advantages of self-attention. Furthermore, there has been limited research into multi-class SAR target detection. To address these limitations, this study proposes a SAR detector named CCDN-DETR, which builds upon the framework of the detection transformer (DETR). To adapt to the multiscale characteristics of SAR data, cross-scale encoders were introduced to facilitate comprehensive information modeling and fusion across different scales. Simultaneously, we optimized the query selection scheme for the input decoder layers, employing IOU loss to assist in initializing object queries more effectively. Additionally, we introduced constrained contrastive denoising training at the decoder layers to enhance the model's convergence speed and improve the detection of different categories of SAR targets. In the benchmark evaluation on a joint dataset composed of SSDD, HRSID, and SAR-AIRcraft datasets, CCDN-DETR achieves a mean Average Precision (mAP) of 91.9%. Furthermore, it demonstrates significant competitiveness with 83.7% mAP on the multi-class MSAR dataset compared to CNN-based models.

A stochastic model based on Gaussian random fields to characterize the morphology of granular objects

The geometrical modeling of granular objects is a complex challenge that exists in many scientific fields, such as the modeling of granular materials or rocks and coarse aggregates with applications in civil, mechanical, and chemical engineering. In this paper, a model called SPHERE (Stochastic Process for Highly Effective Radial Expansion) is proposed, which is based on the deformation of an ellipsoid mesh using multiple 3D Gaussian random fields. The model is designed to be flexible (full control over 2D and 3D morphological properties of granular objects), ultra-fast (over 1000 aggregates in less than 5 s), and independent of the mesh and base shape used (as long as it is a star-shaped object). The flexibility of the model and its ability to reflect real data is illustrated using images of latex nanoparticle aggregates. Using 2D measurements on images from a morphogranulometer, a method based on the SPHERE model is proposed to estimate the 3D morphological properties of aggregates. A multiscale optimization process is applied, in particular using a partial reconstruction of 2D shapes from elliptic Fourier descriptors, in order to best reproduce the shape, angularity and texture of the aggregates using the SPHERE model. Validation of the method on 3D printed data shows relative errors of less than 3% for all measured 2D and 3D morphological characteristics, and validation on a population of synthetic objects shows relative errors of less than 6%. The results are compared and discussed with those obtained using other models based on overlapping spheres and show consistency with previous work. Finally, suggestions for improvement are given.

Transcendentalism and Neuroscience: Grounds for Integration

The author considers the methodological basis of neurotranscendentalism and discusses the question of whether its integration of the approaches of transcendentalism and neurosciences is effective and coherent. The study undertakes on the base of the comparative analysis of spatiotemporal and quantitative representations and how they are described in Kant’s philosophy and transcendental neuro-researches. The author infers that: 1) transcendentalism is not susceptible to the problem of the material correlate of a priori principles of the reason while neurosciences are primarily aimed at detective it; 2) the integrated disciplines have essential diversities in the treatment of the crucial notions (a priori, representation, space, time, magnitude, perception, etc.); 3) neurosciences and transcendentalism have the ground for the integration on the level of the subject area; 4) neurotranscendentalism forms the interdisciplinary scope of researches which generates its own synthetic object – the “biological a priori”.

Synthetic Depth Image-Based Category-Level Object Pose Estimation With Effective Pose Decoupling and Shape Optimization

Synthetic Depth Image-Based Category-Level Object Pose Estimation With Effective Pose Decoupling and Shape Optimization

ОСОБЛИВОСТІ РОЗПОДІЛУ МАКРО- ТА МІКРОПЛАСТИКУ В МЕЖАХ ПЛЯЖЕВИХ ЗОН ВНУТРІШНЬОКОНТИНЕНТАЛЬНИХ МОРІВ (НА ПРИКЛАДІ ПЛЯЖЕВОЇ ЗОНИ М. ЮЖНЕ, УКРАЇНА)

Background. Suburban beaches within the Ukrainian coast of the northwestern shelf of the Black Sea, taking into account the density of industrial and social infrastructure, the proximity to the confluence of large rivers with significant catchment areas into the sea, are indicative of objects for carrying out regulatory studies of the distribution of plastic waste and microplastics in the components of the environment . Methods. The regularities of the distribution of plastic waste and microplastics within the coastal zone of the north-western part of the Black Sea were determined using the example of test sites in the city of Yuzhne (Odesa region, Ukraine). To analyze the quantitative and qualitative characteristics of the synthetic material and to determine the features of its formation, distribution and accumulation in the coastal zone, the method of filtering water flows of the surface runoff was used. Visual analysis of plastic polymers was performed and their types were determined using spectroscopy and OPUS 7.5 software. Results. The results of Raman spectroscopy studies proved that the majority of selected samples of macro- and microplastics are represented by polypropylene, polyethylene and polystyrene. It has been confirmed that the main part of the plastic has a high migration capacity within the coastal areas, which is determined by the transport of runoff during periods of wave activity and the absence of a permanent component in the sand layer within the beach. It was determined that for different dimensional categories of plastic fragments, their species composition changes radically. Polymeric synthetic objects larger than 1 cm are represented mainly by polypropylene, polystyrene, and high-density polyethylene products. In the category of finer dimensions of plastic fragments, both the species spectrum of polymers and their belonging to different economic segments of use is expanding. Synthetic fibers, which make up the majority of objects of the smallest particle size fraction - less than 0.1 mm - are present everywhere and in significant quantitative composition. Conclusions. Research shows that the parameters of the coastline (width, length, slope, shape, presence of vegetation, anthropogenic variability), hydrogen processes and the geological structure of the coastal zone have a significant impact on the formation of the conditions for the accumulation of plastics and microplastics. In order to further study the factors influencing the pollution of the beach area by plastic products and microplastics, monitoring studies are needed within the presented test areas.

Field camera input to virtual phantom (ViP) scanner acquisitions for quality assurance of derived MRI quantities: first implementation and proof-of-principle.

Quality assurance (QA) of measurements derived from MRI can require complicated test phantoms. This work introduces a new QA concept using gradient and transmit RF recordings by a limited field camera (FC) to govern the previous Virtual Phantom (ViP) method. The purpose is to describe the first technical implementation of combined FC+ViP, and illustrate its performance in examples, including quantitative first-pass myocardial perfusion. The new QA concept starts with a synthetic test object (STO) representing some arbitrary test input. Using recordings of the unmodified standard sequence by a gradient and RF waveform camera (FC), ViP calculates by Bloch simulation the continuous RF signal emitted by the STO during this sequence (hence FC+ViP). During nominally identical repetition of the sequence acquisition, ViP transmits the RF signal for scanner reception, reconstruction and any further parametric derivations by the unmodified standard scanner image reconstruction and analysis software. The scanner outputs were compared against the input STOs. First proof-of-principle was discussed and supported by correlation between scanner outputs and the input STO. The work makes no claim that its examples are valid QA methods. It concludes by proposing a new industrial standard for QA without the FC.

Synthetic Detections of Interstellar Objects with the Rubin Observatory Legacy Survey of Space and Time

The discovery of two interstellar objects passing through the solar system, 1I/‘Oumuamua and 2I/Borisov, implies that a galactic population exists with a spatial number density of order ∼0.1 au−3. The forthcoming Rubin Observatory Legacy Survey of Space and Time (LSST) has been predicted to detect more asteroidal interstellar objects like 1I/‘Oumuamua. We apply recently developed methods to simulate a suite of galactic populations of interstellar objects with a range of assumed kinematics, albedos, and size–frequency distributions (SFDs). We incorporate these populations into the objectsInField algorithm, which simulates detections of moving objects by an arbitrary survey. We find that the LSST should detect between ∼0 and 70 asteroidal interstellar objects every year (assuming the implied number density), with sensitive dependence on the SFD slope and characteristic albedo of the host population. The apparent rate of motion on the sky—along with the associated trailing loss—appears to be the largest barrier to detecting interstellar objects. Specifically, a relatively large number of synthetic objects would be detectable by the LSST if not for their rapid sky motion (>0.°5 day−1). Therefore, algorithms that could successfully link and detect rapidly moving objects would significantly increase the number of interstellar object discoveries with the LSST (and in general). The mean diameter of detectable, inactive interstellar objects ranges from ∼50 to 600 m and depends sensitively on the SFD slope and albedo.

Towards robustness and generalization of point cloud representation: A geometry coding method and a large-scale object-level dataset

Robustness and generalization are two challenging problems for learning point cloud representation. To tackle these problems, we first design a novel geometry coding model, which can effectively use an invariant eigengraph to group points with similar geometric information, even when such points are far from each other. We also introduce a large-scale point cloud dataset, PCNet184. It consists of 184 categories and 51,915 synthetic objects, which brings new challenges for point cloud classification, and provides a new benchmark to assess point cloud cross-domain generalization. Finally, we perform extensive experiments on point cloud classification, using ModelNet40, ScanObjectNN, and our PCNet184, and segmentation, using ShapeNetPart and S3DIS. Our method achieves comparable performance to state-of-the-art methods on these datasets, for both supervised and unsupervised learning. Code and our dataset are available at https://github.com/MingyeXu/PCNet184.

Effective selection of shears in variable lateral shearing holography.

The efficiency of reconstruction of complex wavefields in digital holography through shear interferometry has a direct correlation with the shears selected for image acquisition. Although studies to investigate the effect of shears have shown correlations between the selected shear set and the spatial and frequency contents of the reconstructed complex wavefield, to our best knowledge, not much information is available to provide a guide on how to select these shears optimally and what factors to be considered during this selection procedure. In this paper, we study the effect of shear parameters on the phase error through a series of simulations using a synthetic object wavefield and provide a range of shear parameters for optimal reconstruction. Further, we correlated the data by comparing the results with corresponding frequency information density maps.

Neural Impostor: Editing Neural Radiance Fields with Explicit Shape Manipulation

AbstractNeural Radiance Fields (NeRF) have significantly advanced the generation of highly realistic and expressive 3D scenes. However, the task of editing NeRF, particularly in terms of geometry modification, poses a significant challenge. This issue has obstructed NeRF's wider adoption across various applications. To tackle the problem of efficiently editing neural implicit fields, we introduce Neural Impostor, a hybrid representation incorporating an explicit tetrahedral mesh alongside a multigrid implicit field designated for each tetrahedron within the explicit mesh. Our framework bridges the explicit shape manipulation and the geometric editing of implicit fields by utilizing multigrid barycentric coordinate encoding, thus offering a pragmatic solution to deform, composite, and generate neural implicit fields while maintaining a complex volumetric appearance. Furthermore, we propose a comprehensive pipeline for editing neural implicit fields based on a set of explicit geometric editing operations. We show the robustness and adaptability of our system through diverse examples and experiments, including the editing of both synthetic objects and real captured data. Finally, we demonstrate the authoring process of a hybrid synthetic‐captured object utilizing a variety of editing operations, underlining the transformative potential of Neural Impostor in the field of 3D content creation and manipulation.

Probing cosmic homogeneity in the Local Universe

ABSTRACT We investigate the transition scale to homogeneity, RH, using as cosmic tracer the spectroscopic sample of blue galaxies from the Sloan Digital Sky Survey (SDSS). Considering the spatial distribution of the galaxy sample, we compute the two point correlation function ξ(r), the scaled counts in spheres $\mathcal {N}(\lt r)$, and the fractal dimension $\mathcal {D}_2(r)$ to quantify the homogeneity scale in the Local Universe (0.04 < z < 0.20). The sample in analysis is compared with random and mock catalogues with the same geometry, and the same number of synthetic cosmic objects as the data set, to calculate the covariance matrix for the errors determination. The criteria adopted for the transition-to-homogeneity follows the literature, it is attained when $\mathcal {D}_2(r)$ reaches the 1 per cent level of the limit value 3 (i.e. where it reaches 2.97) as the scale increases. We obtain RH = 70.33 ± 10.74 Mpc h−1, at the effective redshift zeff = 0.128, for a sample containing $150\, 302$ SDSS blue galaxies with 0.04 < z < 0.20. Additionally, we perform robustness tests by analysing the homogeneity scale in sub-volumes of the original one, obtaining coherent results; we also check for a possible artefact in our procedure examining a homogeneous synthetic data set as a pseudo-data, verifying that such systematic is absent. Because our analyses concentrate in data at low redshifts, z < 0.20, we find interesting to use cosmography to calculate the radial comoving distances; therefore in this subject our analyses do not use fiducial cosmological model. For completeness, we evaluate the difference of the comoving distances estimation using cosmography and fiducial cosmology.

Standardised images of novel objects created with generative adversarial networks

An enduring question in cognitive science is how perceptually novel objects are processed. Addressing this issue has been limited by the absence of a standardised set of object-like stimuli that appear realistic, but cannot possibly have been previously encountered. To this end, we created a dataset, at the core of which are images of 400 perceptually novel objects. These stimuli were created using Generative Adversarial Networks that integrated features of everyday stimuli to produce a set of synthetic objects that appear entirely plausible, yet do not in fact exist. We curated an accompanying dataset of 400 familiar stimuli, which were matched in terms of size, contrast, luminance, and colourfulness. For each object, we quantified their key visual properties (edge density, entropy, symmetry, complexity, and spectral signatures). We also confirmed that adult observers (N = 390) perceive the novel objects to be less familiar, yet similarly engaging, relative to the familiar objects. This dataset serves as an open resource to facilitate future studies on visual perception.

Shape-from-focus reconstruction using block processing followed by local heat-diffusion-based refinement

Shape from focus (SFF) is a method of recovering the three-dimensional shape of objects from the image sequence through various focusing measurement operators, and is widely used in the fields of medical diagnosis, PCB inspection, industrial manufacturing, and robotic operations. In SFF, the wrong focus measurement causes an inaccurate reconstruction of the object. Generally, to improve the accuracy of focusing measurements, many focus measure (FM) operators are proposed, which is a method that improves the focus volume (FV) performance. However, due to the influence of blur diffusion, which refers to the overlapping of blur circles between different regions, the calculation results of the FM operator in the local window often do not meet the stability requirements of complex scenarios. It is difficult to provide high-precision reconstruction results in practical applications. To overcome this limitation, we proposed an SFF framework based on block processing and heat-diffusion refinement to obtain a refined depth map (RDM). The acquisition of an RDM is mainly divided into two steps. First, the image is divided into blocks to reduce the effect of blur diffusion, and the FM operator is used to calculate the focus measurement value of each block to form the FV. The initial depth map (IDM) is obtained by searching for the best focus position in the FV. The second step is to refine the IDM based on the heat-diffusion principle to obtain the best focus position for all pixels in each block and derive the RDM. The results of experiments conducted on synthetic and real objects demonstrate the superiority of our method in terms of accuracy and its high stability in different application scenarios.

Multiview SVBRDF capture from unified shape and illumination

This paper proposes a stable method for reconstructing spatially varying appearances (SVBRDFs) from multiview images captured under casual lighting conditions. Unlike flat surface capture methods, ours can be applied to surfaces with complex silhouettes. The proposed method takes multiview images as inputs and outputs a unified SVBRDF estimation. We generated a large-scale dataset containing the multiview images, SVBRDFs, and lighting appearance of vast synthetic objects to train a two-stream hierarchical U-Net for SVBRDF estimation that is integrated into a differentiable rendering network for surface appearance reconstruction. In comparison with state-of-the-art approaches, our method produces SVBRDFs with lower biases for more casually captured images.

A Landmark-free Approach for Surface Asymmetry Detection and Profile Drawings from Bilaterally Symmetrical Geometry

Bilaterally symmetrical objects represent a large and important proportion of archaeological artifacts and biological objects. The identification of the plane of symmetry plays a vital role in quantifying surface asymmetry and producing profile drawings in archaeology and anthropology. The correct recognition of symmetry provides evidence to allow experts to restore damaged artifacts, assess consistency in artifact manufacture, and examine morphological variability in human development. With the increasing availability of archaeological and anthropological three-dimensional (3D) meshes, landmark-based and landmark-free morphometric methods for detecting planes of symmetry have both been proposed. However, the landmark-based approach requires manual identification of landmark locations, and hence they are time-consuming and prone to error. Additionally, the landmark-independent morphometric method is influenced by missing data. This study presents an effective landmark-free approach to approximate the best-fitted plane of symmetry from nearly bilaterally symmetrical objects by means of finding the plane with the minimum geometric differences between the original and mirrored meshes. Subsequently, a global and regional method is carried out to quantify surface asymmetry, reducing the effect of the size and orientation of 3D meshes on gross asymmetry detection. Finally, profile drawings are produced by computing the intersections of the plane of symmetry and 3D meshes. Both synthetic and real objects are used to evaluate the effectiveness and robustness of the proposed method. Our results show the approximated plane of symmetry generated by the proposed method is consistent with that determined by anatomical landmarks, and no significant differences in asymmetry ratio representing the degree of gross asymmetry are found between the landmark-based and proposed methods. These results demonstrate that the proposed method provides a suitable plane of symmetry from a bilaterally symmetrical object with small geometric distortion or simple missing geometry, thereby speeding up asymmetry detection and profile drawings.