arXiv subjects

Zahra Dehghanian

Publications and source records attributed to Zahra Dehghanian.

18 recordsLinked to original sources

arXiv2026

ReDiF: Resource-Efficient Few-Step Diffusion Distillation via Reinforcement Learning

Step distillation accelerates diffusion sampling by training a few-step student to imitate a many-step teacher, but distillation itself remains expensive. Typically, this requires thousands of GPU-hours and a large pre-generated trajectory dataset. We introduce ReDiF, which casts step distillation as terminal-reward policy optimization rather than step-wise regression. The student is optimized against a reward computed on the terminal sample, measuring alignment with the teacher's output, instead of matching the teacher's intermediate trajectory under a reconstruction or consistency loss. Because the reward need not be differentiable or trajectory-aligned, ReDiF admits non-differentiable objectives, multi-objective combinations, and preferences the teacher does not express, while exploration lets the student find sampling paths matched to its own step schedule. ReDiF converges in 400 policy updates with 3200 rollouts on a single A100 GPU with 1,000 noise-class pairs and no paired dataset: about 4 GPU-hours, against roughly 336 A100-hours reported for DMD2 on the same EDM teacher. At 8 steps on ImageNet-64, it achieves an FID 3.64 points better than the strongest retrained distillation baseline in the low-training regime under the same single-GPU budget. The formulation is also orthogonal to existing distillation objectives: added to the DMD2 loss, it further improves DMD2's FID by 4.4.

cs.LG↗

arXiv2026

Redefining Generalization in Visual Domains: A Two-Axis Framework for Fake Image Detection with FusionDetect

The rapid development of generative models has made it increasingly crucial to develop detectors that can reliably detect synthetic images. Although most of the work has now focused on cross-generator generalization, we argue that this viewpoint is too limited. Detecting synthetic images involves another equally important challenge: generalization across visual domains. To bridge this gap,we present the OmniGen Benchmark. This comprehensive evaluation dataset incorporates 12 state-of-the-art generators, providing a more realistic way of evaluating detector performance under realistic conditions. In addition, we introduce a new method, FusionDetect, aimed at addressing both vectors of generalization. FusionDetect draws on the benefits of two frozen foundation models: CLIP & Dinov2. By deriving features from both complementary models,we develop a cohesive feature space that naturally adapts to changes in both thecontent and design of the generator. Our extensive experiments demonstrate that FusionDetect delivers not only a new state-of-the-art, which is 3.87% more accurate than its closest competitor and 6.13% more precise on average on established benchmarks, but also achieves a 4.48% increase in accuracy on OmniGen,along with exceptional robustness to common image perturbations. We introduce not only a top-performing detector, but also a new benchmark and framework for furthering universal AI image detection. The code and dataset are available at http://github.com/amir-aman/FusionDetect

cs.CV↗

arXiv2026

Cueless EEG imagined speech for subject identification: dataset and benchmarks

Electroencephalogram (EEG) signals have emerged as a promising modality for biometric identification. While previous studies have explored the use of imagined speech with semantically meaningful words for subject identification, most have relied on additional visual or auditory cues. In this study, we introduce a cueless EEG-based imagined speech paradigm, where subjects imagine the pronunciation of semantically meaningful words without any external cues. This innovative approach addresses the limitations of prior methods by requiring subjects to select and imagine words from a predefined list naturally. The dataset comprises over 4,350 trials from 11 subjects across five sessions. We assess a variety of classification methods, including traditional machine learning techniques such as Support Vector Machines (SVM) and XGBoost, as well as time-series foundation models and deep learning architectures specifically designed for EEG classification, such as EEG Conformer and Shallow ConvNet. A session-based hold-out validation strategy was employed to ensure reliable evaluation and prevent data leakage. Our results demonstrate outstanding classification accuracy, reaching 97.93%. These findings highlight the potential of cueless EEG paradigms for secure and reliable subject identification in real-world applications, such as brain-computer interfaces (BCIs).

cs.LG↗

arXiv2026

MonoVoc: Decoupling Geometry and Semantics for Lightweight Monocular Open-Vocabulary 3D Gaussians

Open vocabulary 3D scene understanding is essential for next-generation interactive systems, empowering users to intuitively query and navigate reconstructed environments using natural language. However, current 3D Gaussian frameworks are often bottlenecked by restrictive multiview capture requirements, costly scene-specific optimization, and the massive memory overhead of storing dense language features. We present a novel, training-free pipeline that fundamentally reimagines this paradigm by explicitly decoupling 3D geometric reconstruction from semantic integration. Given a standard monocular video sequence as input, our method efficiently outputs a compact, highly interpretable, and fully searchable object-level semantic Gaussian map. Rather than entangling heavy language embeddings within the mapping loop, we extract geometry independently and ground semantics through a lightweight, modular post-processing framework. Extensive evaluations on the Replica dataset demonstrate that this decoupled architecture preserves strong rendering fidelity and competitive segmentation accuracy. Crucially, by replacing dense per-Gaussian storage with modular, object-level semantic embeddings, our approach delivers an order-of-magnitude reduction in memory usage compared to SOTA baselines. This provides a highly efficient, scalable, and practical solution for open-vocabulary 3D retrieval and question answering directly from everyday monocular video.

cs.CV↗

arXiv2026

3DLAND: 3D Lesion Abdominal Anomaly Localization Dataset

Existing medical imaging datasets for abdominal CT often lack three-dimensional annotations, multi-organ coverage, or precise lesion-to-organ associations, hindering robust representation learning and clinical applications. To address this gap, we introduce 3DLAND, a large-scale benchmark dataset comprising over 6,000 contrast-enhanced CT volumes with over 20,000 high-fidelity 3D lesion annotations linked to seven abdominal organs: liver, kidneys, pancreas, spleen, stomach, and gallbladder. Our streamlined three-phase pipeline integrates automated spatial reasoning, prompt-optimized 2D segmentation, and memory-guided 3D propagation, validated by expert radiologists with surface dice scores exceeding 0.75. By providing diverse lesion types and patient demographics, 3DLAND enables scalable evaluation of anomaly detection, localization, and cross-organ transfer learning for medical AI. Our dataset establishes a new benchmark for evaluating organ-aware 3D segmentation models, paving the way for advancements in healthcare-oriented AI. To facilitate reproducibility and further research, the 3DLAND dataset and implementation code are publicly available at https://mehrn79.github.io/3DLAND.

eess.IV↗

arXiv2025

CineLOG: A Training Free Approach for Cinematic Long Video Generation

Controllable video synthesis is a central challenge in computer vision, yet current models struggle with fine grained control beyond textual prompts, particularly for cinematic attributes like camera trajectory and genre. Existing datasets often suffer from severe data imbalance, noisy labels, or a significant simulation to real gap. To address this, we introduce CineLOG, a new dataset of 5,000 high quality, balanced, and uncut video clips. Each entry is annotated with a detailed scene description, explicit camera instructions based on a standard cinematic taxonomy, and genre label, ensuring balanced coverage across 17 diverse camera movements and 15 film genres. We also present our novel pipeline designed to create this dataset, which decouples the complex text to video (T2V) generation task into four easier stages with more mature technology. To enable coherent, multi shot sequences, we introduce a novel Trajectory Guided Transition Module that generates smooth spatio-temporal interpolation. Extensive human evaluations show that our pipeline significantly outperforms SOTA end to end T2V models in adhering to specific camera and screenplay instructions, while maintaining professional visual quality. All codes and data are available at https://cine-log.pages.dev.

cs.CV↗

arXiv2025

RealDrag: The First Dragging Benchmark with Real Target Image

The evaluation of drag based image editing models is unreliable due to a lack of standardized benchmarks and metrics. This ambiguity stems from inconsistent evaluation protocols and, critically, the absence of datasets containing ground truth target images, making objective comparisons between competing methods difficult. To address this, we introduce \textbf{RealDrag}, the first comprehensive benchmark for point based image editing that includes paired ground truth target images. Our dataset contains over 400 human annotated samples from diverse video sources, providing source/target images, handle/target points, editable region masks, and descriptive captions for both the image and the editing action. We also propose four novel, task specific metrics: Semantical Distance (SeD), Outer Mask Preserving Score (OMPS), Inner Patch Preserving Score (IPPS), and Directional Similarity (DiS). These metrics are designed to quantify pixel level matching fidelity, check preservation of non edited (out of mask) regions, and measure semantic alignment with the desired task. Using this benchmark, we conduct the first large scale systematic analysis of the field, evaluating 17 SOTA models. Our results reveal clear trade offs among current approaches and establish a robust, reproducible baseline to guide future research. Our dataset and evaluation toolkit will be made publicly available.

cs.CV↗

arXiv2025

Beyond Unified Models: A Service-Oriented Approach to Low Latency, Context Aware Phonemization for Real Time TTS

Lightweight, real-time text-to-speech systems are crucial for accessibility. However, the most efficient TTS models often rely on lightweight phonemizers that struggle with context-dependent challenges. In contrast, more advanced phonemizers with a deeper linguistic understanding typically incur high computational costs, which prevents real-time performance. This paper examines the trade-off between phonemization quality and inference speed in G2P-aided TTS systems, introducing a practical framework to bridge this gap. We propose lightweight strategies for context-aware phonemization and a service-oriented TTS architecture that executes these modules as independent services. This design decouples heavy context-aware components from the core TTS engine, effectively breaking the latency barrier and enabling real-time use of high-quality phonemization models. Experimental results confirm that the proposed system improves pronunciation soundness and linguistic accuracy while maintaining real-time responsiveness, making it well-suited for offline and end-device TTS applications.

cs.SD↗

arXiv2025

MemLoss: Enhancing Adversarial Training with Recycling Adversarial Examples

In this paper, we propose a new approach called MemLoss to improve the adversarial training of machine learning models. MemLoss leverages previously generated adversarial examples, referred to as 'Memory Adversarial Examples,' to enhance model robustness and accuracy without compromising performance on clean data. By using these examples across training epochs, MemLoss provides a balanced improvement in both natural accuracy and adversarial robustness. Experimental results on multiple datasets, including CIFAR-10, demonstrate that our method achieves better accuracy compared to existing adversarial training methods while maintaining strong robustness against attacks.

cs.LG↗

arXiv2025

State of Abdominal CT Datasets: A Critical Review of Bias, Clinical Relevance, and Real-world Applicability

This systematic review critically evaluates publicly available abdominal CT datasets and their suitability for artificial intelligence (AI) applications in clinical settings. We examined 46 publicly available abdominal CT datasets (50,256 studies). Across all 46 datasets, we found substantial redundancy (59.1\% case reuse) and a Western/geographic skew (75.3\% from North America and Europe). A bias assessment was performed on the 19 datasets with >=100 cases; within this subset, the most prevalent high-risk categories were domain shift (63\%) and selection bias (57\%), both of which may undermine model generalizability across diverse healthcare environments -- particularly in resource-limited settings. To address these challenges, we propose targeted strategies for dataset improvement, including multi-institutional collaboration, adoption of standardized protocols, and deliberate inclusion of diverse patient populations and imaging technologies. These efforts are crucial in supporting the development of more equitable and clinically robust AI models for abdominal imaging.

eess.IV↗

arXiv2025

Camera Trajectory Generation: A Comprehensive Survey of Methods, Metrics, and Future Directions

Camera trajectory generation is a cornerstone in computer graphics, robotics, virtual reality, and cinematography, enabling seamless and adaptive camera movements that enhance visual storytelling and immersive experiences. Despite its growing prominence, the field lacks a systematic and unified survey that consolidates essential knowledge and advancements in this domain. This paper addresses this gap by providing the first comprehensive review of the field, covering from foundational definitions to advanced methodologies. We introduce the different approaches to camera representation and present an in-depth review of available camera trajectory generation models, starting with rule-based approaches and progressing through optimization-based techniques, machine learning advancements, and hybrid methods that integrate multiple strategies. Additionally, we gather and analyze the metrics and datasets commonly used for evaluating camera trajectory systems, offering insights into how these tools measure performance, aesthetic quality, and practical applicability. Finally, we highlight existing limitations, critical gaps in current research, and promising opportunities for investment and innovation in the field. This paper not only serves as a foundational resource for researchers entering the field but also paves the way for advancing adaptive, efficient, and creative camera trajectory systems across diverse applications.

cs.CV↗

arXiv2025

LensCraft: Your Professional Virtual Cinematographer

Digital creators, from indie filmmakers to animation studios, face a persistent bottleneck: translating their creative vision into precise camera movements. Despite significant progress in computer vision and artificial intelligence, current automated filming systems struggle with a fundamental trade-off between mechanical execution and creative intent. Crucially, almost all previous works simplify the subject to a single point-ignoring its orientation and true volume-severely limiting spatial awareness during filming. LensCraft solves this problem by mimicking the expertise of a professional cinematographer, using a data-driven approach that combines cinematographic principles with the flexibility to adapt to dynamic scenes in real time. Our solution combines a specialized simulation framework for generating high-fidelity training data with an advanced neural model that is faithful to the script while being aware of the volume and dynamic behavior of the subject. Additionally, our approach allows for flexible control via various input modalities, including text prompts, subject trajectory and volume, key points, or a full camera trajectory, offering creators a versatile tool to guide camera movements in line with their vision. Leveraging a lightweight real time architecture, LensCraft achieves markedly lower computational complexity and faster inference while maintaining high output quality. Extensive evaluation across static and dynamic scenarios reveals unprecedented accuracy and coherence, setting a new benchmark for intelligent camera systems compared to state-of-the-art models. Extended results, the complete dataset, simulation environment, trained model weights, and source code are publicly accessible on LensCraft Webpage.

cs.GR↗

arXiv2025

Fast, Not Fancy: Rethinking G2P with Rich Data and Rule-Based Models

Homograph disambiguation remains a significant challenge in grapheme-to-phoneme (G2P) conversion, especially for low-resource languages. This challenge is twofold: (1) creating balanced and comprehensive homograph datasets is labor-intensive and costly, and (2) specific disambiguation strategies introduce additional latency, making them unsuitable for real-time applications such as screen readers and other accessibility tools. In this paper, we address both issues. First, we propose a semi-automated pipeline for constructing homograph-focused datasets, introduce the HomoRich dataset generated through this pipeline, and demonstrate its effectiveness by applying it to enhance a state-of-the-art deep learning-based G2P system for Persian. Second, we advocate for a paradigm shift - utilizing rich offline datasets to inform the development of fast, rule-based methods suitable for latency-sensitive accessibility applications like screen readers. To this end, we improve one of the most well-known rule-based G2P systems, eSpeak, into a fast homograph-aware version, HomoFast eSpeak. Our results show an approximate 30% improvement in homograph disambiguation accuracy for the deep learning-based and eSpeak systems.

cs.CL↗

arXiv2025

UPL: Uncertainty-aware Pseudo-labeling for Imbalance Transductive Node Classification

Graph-structured datasets often suffer from class imbalance, which complicates node classification tasks. In this work, we address this issue by first providing an upper bound on population risk for imbalanced transductive node classification. We then propose a simple and novel algorithm, Uncertainty-aware Pseudo-labeling (UPL). Our approach leverages pseudo-labels assigned to unlabeled nodes to mitigate the adverse effects of imbalance on classification accuracy. Furthermore, the UPL algorithm enhances the accuracy of pseudo-labeling by reducing training noise of pseudo-labels through a novel uncertainty-aware approach. We comprehensively evaluate the UPL algorithm across various benchmark datasets, demonstrating its superior performance compared to existing state-of-the-art methods.

cs.LG↗

arXiv2025

Beyond Imperfections: A Conditional Inpainting Approach for End-to-End Artifact Removal in VTON and Pose Transfer

Artifacts often degrade the visual quality of virtual try-on (VTON) and pose transfer applications, impacting user experience. This study introduces a novel conditional inpainting technique designed to detect and remove such distortions, improving image aesthetics. Our work is the first to present an end-to-end framework addressing this specific issue, and we developed a specialized dataset of artifacts in VTON and pose transfer tasks, complete with masks highlighting the affected areas. Experimental results show that our method not only effectively removes artifacts but also significantly enhances the visual quality of the final images, setting a new benchmark in computer vision and image processing.

cs.CV↗

arXiv2024

LLM-Powered Grapheme-to-Phoneme Conversion: Benchmark and Case Study

Grapheme-to-phoneme (G2P) conversion is critical in speech processing, particularly for applications like speech synthesis. G2P systems must possess linguistic understanding and contextual awareness of languages with polyphone words and context-dependent phonemes. Large language models (LLMs) have recently demonstrated significant potential in various language tasks, suggesting that their phonetic knowledge could be leveraged for G2P. In this paper, we evaluate the performance of LLMs in G2P conversion and introduce prompting and post-processing methods that enhance LLM outputs without additional training or labeled data. We also present a benchmarking dataset designed to assess G2P performance on sentence-level phonetic challenges of the Persian language. Our results show that by applying the proposed methods, LLMs can outperform traditional G2P tools, even in an underrepresented language like Persian, highlighting the potential of developing LLM-aided G2P systems.

cs.CL↗

arXiv2024

ManaTTS Persian: a recipe for creating TTS datasets for lower resource languages

In this study, we introduce ManaTTS, the most extensive publicly accessible single-speaker Persian corpus, and a comprehensive framework for collecting transcribed speech datasets for the Persian language. ManaTTS, released under the open CC-0 license, comprises approximately 86 hours of audio with a sampling rate of 44.1 kHz. Alongside ManaTTS, we also generated the VirgoolInformal dataset to evaluate Persian speech recognition models used for forced alignment, extending over 5 hours of audio. The datasets are supported by a fully transparent, MIT-licensed pipeline, a testament to innovation in the field. It includes unique tools for sentence tokenization, bounded audio segmentation, and a novel forced alignment method. This alignment technique is specifically designed for low-resource languages, addressing a crucial need in the field. With this dataset, we trained a Tacotron2-based TTS model, achieving a Mean Opinion Score (MOS) of 3.76, which is remarkably close to the MOS of 3.86 for the utterances generated by the same vocoder and natural spectrogram, and the MOS of 4.01 for the natural waveform, demonstrating the exceptional quality and effectiveness of the corpus.

cs.SD↗

arXiv2024

Spot The Odd One Out: Regularized Complete Cycle Consistent Anomaly Detector GAN

This study presents an adversarial method for anomaly detection in real-world applications, leveraging the power of generative adversarial neural networks (GANs) through cycle consistency in reconstruction error. Previous methods suffer from the high variance between class-wise accuracy which leads to not being applicable for all types of anomalies. The proposed method named RCALAD tries to solve this problem by introducing a novel discriminator to the structure, which results in a more efficient training process. Additionally, RCALAD employs a supplementary distribution in the input space to steer reconstructions toward the normal data distribution, effectively separating anomalous samples from their reconstructions and facilitating more accurate anomaly detection. To further enhance the performance of the model, two novel anomaly scores are introduced. The proposed model has been thoroughly evaluated through extensive experiments on six various datasets, yielding results that demonstrate its superiority over existing state-of-the-art models. The code is readily available to the research community at https://github.com/zahraDehghanian97/RCALAD.

cs.LG↗