Dimitris Samaras
YOU?
Author Swipe
View article: Reward Modeling of Goal-directed Gaze Control
Reward Modeling of Goal-directed Gaze Control Open
Goal-directed visual search in natural scenes is a complex behavior that requires the flexible integration of vision, memory, and contextual knowledge. Here we introduce a reward-based framework that unifies these processes by learning tar…
View article: Multi-view Gaze Target Estimation
Multi-view Gaze Target Estimation Open
This paper presents a method that utilizes multiple camera views for the gaze target estimation (GTE) task. The approach integrates information from different camera views to improve accuracy and expand applicability, addressing limitation…
View article: Low-Rank Head Avatar Personalization with Registers
Low-Rank Head Avatar Personalization with Registers Open
We introduce a novel method for low-rank personalization of a generic model for head avatar generation. Prior work proposes generic models that achieve high-quality face animation by leveraging large-scale datasets of multiple identities. …
View article: Adaptive Multitask Neural Network for High-Fidelity Wake Flow Modeling of Wind Farms
Adaptive Multitask Neural Network for High-Fidelity Wake Flow Modeling of Wind Farms Open
Wind turbine wake modeling is critical for the design and optimization of wind farms. Traditional methods often struggle with the trade-off between accuracy and computational cost. Recently, data-driven neural networks have emerged as a pr…
View article: Ciliated cell domains with locally coordinated ciliary motion generate a mosaic of microflows in the brain’s lateral ventricles
Ciliated cell domains with locally coordinated ciliary motion generate a mosaic of microflows in the brain’s lateral ventricles Open
Circulation of cerebrospinal fluid (CSF) through the brain’s ventricles is essential for maintaining brain homeostasis and supporting neurogenesis. CSF flow is supported by the structural polarization of multiciliated cells, which align wi…
View article: TopoCellGen: Generating Histopathology Cell Topology with a Diffusion Model
TopoCellGen: Generating Histopathology Cell Topology with a Diffusion Model Open
Accurately modeling multi-class cell topology is crucial in digital pathology, as it provides critical insights into tissue structure and pathology. The synthetic generation of cell topology enables realistic simulations of complex tissue …
View article: MLI-NeRF: Multi-Light Intrinsic-Aware Neural Radiance Fields
MLI-NeRF: Multi-Light Intrinsic-Aware Neural Radiance Fields Open
Current methods for extracting intrinsic image components, such as reflectance and shading, primarily rely on statistical priors. These methods focus mainly on simple synthetic scenes and isolated objects and struggle to perform well on ch…
View article: MLI-NeRF: Multi-Light Intrinsic-Aware Neural Radiance Fields
MLI-NeRF: Multi-Light Intrinsic-Aware Neural Radiance Fields Open
Current methods for extracting intrinsic image components, such as reflectance and shading, primarily rely on statistical priors. These methods focus mainly on simple synthetic scenes and isolated objects and struggle to perform well on ch…
View article: Instance-Aware Generalized Referring Expression Segmentation
Instance-Aware Generalized Referring Expression Segmentation Open
Recent works on Generalized Referring Expression Segmentation (GRES) struggle with handling complex expressions referring to multiple distinct objects. This is because these methods typically employ an end-to-end foreground-background segm…
View article: Direct and Explicit 3D Generation from a Single Image
Direct and Explicit 3D Generation from a Single Image Open
Current image-to-3D approaches suffer from high computational costs and lack scalability for high-resolution outputs. In contrast, we introduce a novel framework to directly generate explicit surface geometry and texture using multi-view 2…
View article: Fast constrained sampling in pre-trained diffusion models
Fast constrained sampling in pre-trained diffusion models Open
Large denoising diffusion models, such as Stable Diffusion, have been trained on billions of image-caption pairs to perform text-conditioned image generation. As a byproduct of this training, these models have acquired general knowledge ab…
View article: TopoDiffusionNet: A Topology-aware Diffusion Model
TopoDiffusionNet: A Topology-aware Diffusion Model Open
Diffusion models excel at creating visually impressive images but often struggle to generate images with a specified topology. The Betti number, which represents the number of structures in an image, is a fundamental measure in topology. Y…
View article: Learning Frame-Wise Emotion Intensity for Audio-Driven Talking-Head Generation
Learning Frame-Wise Emotion Intensity for Audio-Driven Talking-Head Generation Open
Human emotional expression is inherently dynamic, complex, and fluid, characterized by smooth transitions in intensity throughout verbal communication. However, the modeling of such intensity fluctuations has been largely overlooked by pre…
View article: JEAN: Joint Expression and Audio-guided NeRF-based Talking Face Generation
JEAN: Joint Expression and Audio-guided NeRF-based Talking Face Generation Open
We introduce a novel method for joint expression and audio-guided talking face generation. Recent approaches either struggle to preserve the speaker identity or fail to produce faithful facial expressions. To address these challenges, we p…
View article: Shadow Removal Refinement via Material-Consistent Shadow Edges
Shadow Removal Refinement via Material-Consistent Shadow Edges Open
Shadow boundaries can be confused with material boundaries as both exhibit sharp changes in luminance or contrast within a scene. However, shadows do not modify the intrinsic color or texture of surfaces. Therefore, on both sides of shadow…
View article: Look Hear: Gaze Prediction for Speech-directed Human Attention
Look Hear: Gaze Prediction for Speech-directed Human Attention Open
For computer systems to effectively interact with humans using spoken language, they need to understand how the words being generated affect the users' moment-by-moment attention. Our study focuses on the incremental prediction of attentio…
View article: Assessing Sample Quality via the Latent Space of Generative Models
Assessing Sample Quality via the Latent Space of Generative Models Open
Advances in generative models increase the need for sample quality assessment. To do so, previous methods rely on a pre-trained feature extractor to embed the generated samples and real samples into a common space for comparison. However, …
View article: MIGS: Multi-Identity Gaussian Splatting via Tensor Decomposition
MIGS: Multi-Identity Gaussian Splatting via Tensor Decomposition Open
We introduce MIGS (Multi-Identity Gaussian Splatting), a novel method that learns a single neural representation for multiple identities, using only monocular videos. Recent 3D Gaussian Splatting (3DGS) approaches for human avatars require…
View article: Uncertainty Estimation for Tumor Prediction with Unlabeled Data
Uncertainty Estimation for Tumor Prediction with Unlabeled Data Open
Estimating uncertainty of a neural network is crucial in providing transparency and trustworthiness. In this paper, we focus on uncertainty estimation for digital pathology prediction models. To explore the large amount of unlabeled data i…
View article: Learning Relighting and Intrinsic Decomposition in Neural Radiance Fields
Learning Relighting and Intrinsic Decomposition in Neural Radiance Fields Open
The task of extracting intrinsic components, such as reflectance and shading, from neural radiance fields is of growing interest. However, current methods largely focus on synthetic scenes and isolated objects, overlooking the complexities…
View article: Learned Representation-Guided Diffusion Models for Large-Image Generation
Learned Representation-Guided Diffusion Models for Large-Image Generation Open
To synthesize high-fidelity samples, diffusion models typically require auxiliary data to guide the generation process. However, it is impractical to procure the painstaking patch-level annotation effort required in specialized domains lik…
View article: Diffusion-Refined VQA Annotations for Semi-Supervised Gaze Following
Diffusion-Refined VQA Annotations for Semi-Supervised Gaze Following Open
Training gaze following models requires a large number of images with gaze target coordinates annotated by human annotators, which is a laborious and inherently ambiguous process. We propose the first semi-supervised method for gaze follow…
View article: Toward ultra-efficient high fidelity predictions of wind turbine wakes: Augmenting the accuracy of engineering models via LES-trained machine learning
Toward ultra-efficient high fidelity predictions of wind turbine wakes: Augmenting the accuracy of engineering models via LES-trained machine learning Open
This study proposes a novel machine learning (ML) methodology for the efficient and cost-effective prediction of high-fidelity three-dimensional velocity fields in the wake of utility-scale turbines. The model consists of an auto-encoder c…
View article: MI-NeRF: Learning a Single Face NeRF from Multiple Identities
MI-NeRF: Learning a Single Face NeRF from Multiple Identities Open
In this work, we introduce a method that learns a single dynamic neural radiance field (NeRF) from monocular talking face videos of multiple identities. NeRFs have shown remarkable results in modeling the 4D dynamics and appearance of huma…
View article: Self-supervised co-salient object detection via feature correspondence at multiple scales
Self-supervised co-salient object detection via feature correspondence at multiple scales Open
Our paper introduces a novel two-stage self-supervised approach for detecting co-occurring salient objects (CoSOD) in image groups without requiring segmentation annotations. Unlike existing unsupervised methods that rely solely on patch-l…
View article: Rig3DGS: Creating Controllable Portraits from Casual Monocular Videos
Rig3DGS: Creating Controllable Portraits from Casual Monocular Videos Open
Creating controllable 3D human portraits from casual smartphone videos is highly desirable due to their immense value in AR/VR applications. The recent development of 3D Gaussian Splatting (3DGS) has shown improvements in rendering quality…
View article: SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology
SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology Open
Introducing interpretability and reasoning into Multiple Instance Learning (MIL) methods for Whole Slide Image (WSI) analysis is challenging, given the complexity of gigapixel slides. Traditionally, MIL interpretability is limited to ident…
View article: MORCIC: Model Order Reduction Techniques for Electromagnetic Models of Integrated Circuits
MORCIC: Model Order Reduction Techniques for Electromagnetic Models of Integrated Circuits Open
Model order reduction (MOR) is crucial for the design process of integrated circuits. Specifically, the vast amount of passive RLCk elements in electromagnetic models extracted from physical layouts exacerbates the extraction time, the sto…
View article: Unsupervised and semi-supervised co-salient object detection via segmentation frequency statistics
Unsupervised and semi-supervised co-salient object detection via segmentation frequency statistics Open
In this paper, we address the detection of co-occurring salient objects (CoSOD) in an image group using frequency statistics in an unsupervised manner, which further enable us to develop a semi-supervised method. While previous works have …
View article: A systematic study of key elements underlying molecular property prediction
A systematic study of key elements underlying molecular property prediction Open