arXiv subjects

Marcelo dos Santos

Publications and source records attributed to Marcelo dos Santos.

8 recordsLinked to original sources

arXiv2026

Robust Face Super-Resolution and Recognition Through Multi-Feature Aggregation in Diffusion Models

Images acquired in surveillance environments often suffer from conditions such as low resolution, variations in pose, irregular illumination, and occlusions. Due to the low quality of these images, face recognition algorithms often struggle. This major limitation can be addressed by employing super-resolution techniques that enhance the details of the image. However, due to the high degree of difficulty of the problem, most super-resolution algorithms tend to cause distortions in the image and in the individual's identity. Thus, additional information must be incorporated into the processing to improve recognition robustness. In this regard, surveillance cameras can capture multiple images, even at low quality, and the data extracted from these images, such as consecutive video frames, can significantly enhance both super-resolution and facial recognition. In this work, we introduce FASR++, a diffusion-model-based super-resolution algorithm. It leverages a reference low-resolution image and features extracted from multiple auxiliary low-quality images to generate a super-resolved output, minimizing distortions in the individual's identity. Our approach recovers facial features without explicitly providing soft attributes or computing a function gradient to guide the reconstruction process. FASR++ generates high-quality images that can considerably improve performance in face recognition tasks when used as a pre-processing step. We validate our approach on two standard face recognition datasets and attain state-of-the-art results for verification, face recognition, and image quality metrics such as PSNR, SSIM, and LPIPS.

cs.CV↗

arXiv2026

The 11D pure spinor ghost number zero vertex operator

The 11D pure spinor worldline has been proved to successfully describe the physical states of 11D supergravity in a manifestly super-Poincaré covariant fashion. Within this framework, the computation of scattering amplitudes requires the existence of vertex operators carrying different ghost numbers. A recent no-go theorem demonstrated the impossibility of constructing a ghost number zero vertex operator consistent with 11D supergravity in the minimal pure spinor formalism. In this letter, we overcome this obstruction by working in the non-minimal formulation of the 11D pure spinor superparticle. We construct, for the first time, a ghost number zero vertex operator with a remarkably compact structure when expressed in terms of physical operators. We further verify that it satisfies the expected descent relation with the ghost number one vertex operator, and that its commutator with the ghost number three single-particle vertex reproduces the two-particle superfield recently introduced in the literature.

hep-th↗

arXiv2025

The pure spinor superparticle and 10D super-Yang-Mills amplitudes

We present a prescription for computing tree-level scattering amplitudes in 10D super-Yang-Mills (SYM) theory using the pure spinor worldline formalism. The pure spinor formalism has proven to be a powerful framework for studying supersymmetric field theories, providing manifestly covariant and BRST-invariant formulations of amplitudes. By incorporating the worldline approach, we construct a first-quantized representation of SYM amplitudes in 10D, where interactions are encoded through the insertion of vertex operators along the particle's trajectory. We explicitly compute the N-point function, demonstrating an agreement with the limit $α' \to 0$ of open superstring amplitudes and confirming that the kinematic numerators satisfy the expected BRST relations. Our results establish the pure spinor worldline formalism as a tool for studying scattering amplitudes and suggest further applications to 11D supergravity.

hep-th↗

arXiv2025

Tree-level 11D supergravity amplitudes from the pure spinor worldline

We develop a pure spinor worldline formalism for computing tree-level scattering amplitudes in 11D supergravity. Focusing first on the 4-point amplitude, we demonstrate that our prescription is consistent with BRST symmetry and gauge invariance, and that the resulting expression is invariant under permutation of the external particles. Remarkably, the amplitude admits a compact representation in pure spinor superspace and agrees precisely with the result obtained via perturbiner methods. We further extend our construction to the N-point case, proposing a general correlator that preserves BRST closure and gauge invariance, thereby offering a systematic framework for higher-point computations in 11D supergravity.

hep-th↗

arXiv2025

Improving Small Drone Detection Through Multi-Scale Processing and Data Augmentation

Detecting small drones, often indistinguishable from birds, is crucial for modern surveillance. This work introduces a drone detection methodology built upon the medium-sized YOLOv11 object detection model. To enhance its performance on small targets, we implemented a multi-scale approach in which the input image is processed both as a whole and in segmented parts, with subsequent prediction aggregation. We also utilized a copy-paste data augmentation technique to enrich the training dataset with diverse drone and bird examples. Finally, we implemented a post-processing technique that leverages frame-to-frame consistency to mitigate missed detections. The proposed approach attained first place in the 8th WOSDETC Drone-vs-Bird Detection Grand Challenge, held at the 2025 International Joint Conference on Neural Networks (IJCNN), showcasing its capability to detect drones in complex environments effectively.

cs.CV↗

arXiv2024

Toward Enhancing Vehicle Color Recognition in Adverse Conditions: A Dataset and Benchmark

Vehicle information recognition is crucial in various practical domains, particularly in criminal investigations. Vehicle Color Recognition (VCR) has garnered significant research interest because color is a visually distinguishable attribute of vehicles and is less affected by partial occlusion and changes in viewpoint. Despite the success of existing methods for this task, the relatively low complexity of the datasets used in the literature has been largely overlooked. This research addresses this gap by compiling a new dataset representing a more challenging VCR scenario. The images - sourced from six license plate recognition datasets - are categorized into eleven colors, and their annotations were validated using official vehicle registration information. We evaluate the performance of four deep learning models on a widely adopted dataset and our proposed dataset to establish a benchmark. The results demonstrate that our dataset poses greater difficulty for the tested models and highlights scenarios that require further exploration in VCR. Remarkably, nighttime scenes account for a significant portion of the errors made by the best-performing model. This research provides a foundation for future studies on VCR, while also offering valuable insights for the field of fine-grained vehicle classification.

cs.CV↗

arXiv2022

Face Super-Resolution Using Stochastic Differential Equations

Diffusion models have proven effective for various applications such as images, audio and graph generation. Other important applications are image super-resolution and the solution of inverse problems. More recently, some works have used stochastic differential equations (SDEs) to generalize diffusion models to continuous time. In this work, we introduce SDEs to generate super-resolution face images. To the best of our knowledge, this is the first time SDEs have been used for such an application. The proposed method provides an improved peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and consistency than the existing super-resolution methods based on diffusion models. In particular, we also assess the potential application of this method for the face recognition task. A generic facial feature extractor is used to compare the super-resolution images with the ground truth and superior results were obtained compared with other methods. Our code is publicly available at https://github.com/marcelowds/sr-sde

cs.CV↗

arXiv2011

Sonine approximation for collisional moments of granular gases of inelastic rough spheres

We consider a dilute granular gas of hard spheres colliding inelastically with coefficients of normal and tangential restitution $α$ and $β$, respectively. The basic quantities characterizing the distribution function $f(\mathbf{v},\bmω)$ of linear ($\mathbf{v}$) and angular ($\bmω$) velocities are the second-degree moments defining the translational ($T^\text{tr}$) and rotational ($T^\text{rot}$) temperatures. The deviation of $f$ from the Maxwellian distribution parameterized by $T^\text{tr}$ and $T^\text{rot}$ can be measured by the cumulants associated with the fourth-degree velocity moments. The main objective of this paper is the evaluation of the collisional rates of change of these second- and fourth-degree moments by means of a Sonine approximation. The results are subsequently applied to the computation of the temperature ratio $T^\text{rot}/T^\text{tr}$ and the cumulants of two paradigmatic states: the homogeneous cooling state and the homogeneous steady state driven by a white-noise stochastic thermostat. It is found in both cases that the Maxwellian approximation for the temperature ratio does not deviate much from the Sonine prediction. On the other hand, non-Maxwellian properties measured by the cumulants cannot be ignored, especially in the homogeneous cooling state for medium and small roughness. In that state, moreover, the cumulant directly related to the translational velocity differs in the quasi-smooth limit $β\to -1$ from that of pure smooth spheres ($β=-1$). This singular behavior is directly related to the unsteady character of the homogeneous cooling state and thus it is absent in the stochastic thermostat case.

cond-mat.soft↗