arXiv subjects

Florian Schiffers

Publications and source records attributed to Florian Schiffers.

At least 19 recordsLinked to original sources

arXiv2026

CounterVid: Counterfactual Video Generation for Mitigating Action and Temporal Hallucinations in Video-Language Models

Video-language models (VLMs) achieve strong multimodal understanding but remain prone to hallucinations, especially when reasoning about actions and temporal order. Existing mitigation strategies, such as textual filtering or random video perturbations, often fail to address the root cause: over-reliance on language priors rather than fine-grained visual dynamics. We propose a scalable framework for counterfactual video generation that synthesizes videos differing only in actions or temporal structure while preserving scene context. Our pipeline combines multimodal LLMs for action proposal and editing guidance with diffusion-based image and video models to generate semantic hard negatives at scale. Using this framework, we build CounterVid, a synthetic dataset of ~26k preference pairs constructed from short counterfactual action clips and targeting both action recognition and controlled action-sequence ordering. We further introduce MixDPO, a unified Direct Preference Optimization approach that jointly leverages textual and visual preferences. Across Qwen2.5-VL and InternVL3 backbones, MixDPO substantially improves action recognition and temporal ordering, yields gains on most standard video hallucination benchmarks, and largely preserves general video understanding. Our source code, trained models, and dataset are available at https://aimagelab.github.io/CounterVid.

cs.CV↗

arXiv2026

Do Models Share Safety Representations? Cross-Model Steering for Safe Visual Generation

Recent progress in generative modeling has made safety control a central challenge, yet existing approaches remain largely model-specific, requiring retraining or tailored interventions for each new architecture. In this work, we ask whether safety can be represented as a portable latent direction, learned once and reused across heterogeneous generators. We introduce the first framework for cross-model safety steering, in which a safety direction is estimated in a source LLM from paired safe-unsafe prompts, transported to a target generator through a lightweight alignment fitted on benign data alone, and applied at inference time. Crucially, our pipeline never accesses unsafe data on the target side, isolating whether safety can be transferred through shared representation geometry. Beyond a single global direction, we also identify a multi-vector extension that captures category-specific safety behaviors, enabling more selective control. We evaluate our approach in text-to-image and text-to-video generation across diverse source-target model pairs. Across models, transferred safety directions achieve ASR reduction and CLIP-Score/FID trade-offs comparable to directions learned natively on the target model using unsafe data, while requiring no target-side unsafe data. This indicates that safety improvements do not come at the expense of generation quality. Our results point to a modular view of safety: safety-relevant behavior is not purely model-local, but can be controlled through latent directions that persist across models. This suggests a new path toward lightweight, reusable safety mechanisms that do not require target-side unsafe data.

cs.CV↗

arXiv2025

High contrast holography through dual modulation

Holographic displays are a promising technology for immersive visual experiences, and their potential for compact form factor makes them a strong candidate for head-mounted displays. However, at the short propagation distances needed for a compact, head-mounted architecture, image contrast is low when using a traditional phase-only spatial light modulator (SLM). Although a complex SLM could restore contrast, these modulators require bulky lenses to optically co-locate the amplitude and phase components, making them poorly suited for a compact head-mounted design. In this work, we introduce a novel architecture to improve contrast: by adding a low resolution amplitude SLM a short distance away from the phase modulator, we demonstrate peak signal-to-noise ratio improvement up to 31 dB in simulation and 6.5 dB experimentally compared to phase-only modulation, even when the amplitude modulator is 60$\times$ lower resolution than its phase counterpart. We analyze the relationship between diffraction angle and amplitude modulator pixel size, and validate the concept with a benchtop experimental prototype. By showing that low resolution modulation is sufficient to improve contrast, we open new design spaces for high-contrast holographic displays.

physics.optics↗

arXiv2025

What Happens Next? Next Scene Prediction with a Unified Video Model

Recent unified models for joint understanding and generation have significantly advanced visual generation capabilities. However, their focus on conventional tasks like text-to-video generation has left the temporal reasoning potential of unified models largely underexplored. To address this gap, we introduce Next Scene Prediction (NSP), a new task that pushes unified video models toward temporal and causal reasoning. Unlike text-to-video generation, NSP requires predicting plausible futures from preceding context, demanding deeper understanding and reasoning. To tackle this task, we propose a unified framework combining Qwen-VL for comprehension and LTX for synthesis, bridged by a latent query embedding and a connector module. This model is trained in three stages on our newly curated, large-scale NSP dataset: text-to-video pre-training, supervised fine-tuning, and reinforcement learning (via GRPO) with our proposed causal consistency reward. Experiments demonstrate our model achieves state-of-the-art performance on our benchmark, advancing the capability of generalist multimodal systems to anticipate what happens next.

cs.CV↗

arXiv2025

ScarNet: A Novel Foundation Model for Automated Myocardial Scar Quantification from LGE in Cardiac MRI

Background: Late Gadolinium Enhancement (LGE) imaging is the gold standard for assessing myocardial fibrosis and scarring, with left ventricular (LV) LGE extent predicting major adverse cardiac events (MACE). Despite its importance, routine LGE-based LV scar quantification is hindered by labor-intensive manual segmentation and inter-observer variability. Methods: We propose ScarNet, a hybrid model combining a transformer-based encoder from the Medical Segment Anything Model (MedSAM) with a convolution-based U-Net decoder, enhanced by tailored attention blocks. ScarNet was trained on 552 ischemic cardiomyopathy patients with expert segmentations of myocardial and scar boundaries and tested on 184 separate patients. Results: ScarNet achieved robust scar segmentation in 184 test patients, yielding a median Dice score of 0.912 (IQR: 0.863--0.944), significantly outperforming MedSAM (median Dice = 0.046, IQR: 0.043--0.047) and nnU-Net (median Dice = 0.638, IQR: 0.604--0.661). ScarNet demonstrated lower bias (-0.63%) and coefficient of variation (4.3%) compared to MedSAM (bias: -13.31%, CoV: 130.3%) and nnU-Net (bias: -2.46%, CoV: 20.3%). In Monte Carlo simulations with noise perturbations, ScarNet achieved significantly higher scar Dice (0.892 \pm 0.053, CoV = 5.9%) than MedSAM (0.048 \pm 0.112, CoV = 233.3%) and nnU-Net (0.615 \pm 0.537, CoV = 28.7%). Conclusion: ScarNet outperformed MedSAM and nnU-Net in accurately segmenting myocardial and scar boundaries in LGE images. The model exhibited robust performance across diverse image qualities and scar patterns.

eess.IV↗

arXiv2024

HoloChrome: Polychromatic Illumination for Speckle Reduction in Holographic Near-Eye Displays

Holographic displays hold the promise of providing authentic depth cues, resulting in enhanced immersive visual experiences for near-eye applications. However, current holographic displays are hindered by speckle noise, which limits accurate reproduction of color and texture in displayed images. We present HoloChrome, a polychromatic holographic display framework designed to mitigate these limitations. HoloChrome utilizes an ultrafast, wavelength-adjustable laser and a dual-Spatial Light Modulator (SLM) architecture, enabling the multiplexing of a large set of discrete wavelengths across the visible spectrum. By leveraging spatial separation in our dual-SLM setup, we independently manipulate speckle patterns across multiple wavelengths. This novel approach effectively reduces speckle noise through incoherent averaging achieved by wavelength multiplexing. Our method is complementary to existing speckle reduction techniques, offering a new pathway to address this challenge. Furthermore, the use of polychromatic illumination broadens the achievable color gamut compared to traditional three-color primary holographic displays. Our simulations and tabletop experiments validate that HoloChrome significantly reduces speckle noise and expands the color gamut. These advancements enhance the performance of holographic near-eye displays, moving us closer to practical, immersive next-generation visual experiences.

cs.GR↗

arXiv2024

DRL-STNet: Unsupervised Domain Adaptation for Cross-modality Medical Image Segmentation via Disentangled Representation Learning

Unsupervised domain adaptation (UDA) is essential for medical image segmentation, especially in cross-modality data scenarios. UDA aims to transfer knowledge from a labeled source domain to an unlabeled target domain, thereby reducing the dependency on extensive manual annotations. This paper presents DRL-STNet, a novel framework for cross-modality medical image segmentation that leverages generative adversarial networks (GANs), disentangled representation learning (DRL), and self-training (ST). Our method leverages DRL within a GAN to translate images from the source to the target modality. Then, the segmentation model is initially trained with these translated images and corresponding source labels and then fine-tuned iteratively using a combination of synthetic and real images with pseudo-labels and real labels. The proposed framework exhibits superior performance in abdominal organ segmentation on the FLARE challenge dataset, surpassing state-of-the-art methods by 11.4% in the Dice similarity coefficient and by 13.1% in the Normalized Surface Dice metric, achieving scores of 74.21% and 80.69%, respectively. The average running time is 41 seconds, and the area under the GPU memory-time curve is 11,292 MB. These results indicate the potential of DRL-STNet for enhancing cross-modality medical image segmentation tasks.

eess.IV↗

arXiv2024

Neural Étendue Expander for Ultra-Wide-Angle High-Fidelity Holographic Display

Holographic displays can generate light fields by dynamically modulating the wavefront of a coherent beam of light using a spatial light modulator, promising rich virtual and augmented reality applications. However, the limited spatial resolution of existing dynamic spatial light modulators imposes a tight bound on the diffraction angle. As a result, modern holographic displays possess low étendue, which is the product of the display area and the maximum solid angle of diffracted light. The low étendue forces a sacrifice of either the field-of-view (FOV) or the display size. In this work, we lift this limitation by presenting neural étendue expanders. This new breed of optical elements, which is learned from a natural image dataset, enables higher diffraction angles for ultra-wide FOV while maintaining both a compact form factor and the fidelity of displayed contents to human viewers. With neural étendue expanders, we experimentally achieve 64$\times$ étendue expansion of natural images in full color, expanding the FOV by an order of magnitude horizontally and vertically, with high-fidelity reconstruction quality (measured in PSNR) over 29 dB on retinal-resolution images.

eess.IV↗

arXiv2023

Simultaneous Color Computer Generated Holography

Computer generated holography has long been touted as the future of augmented and virtual reality (AR/VR) displays, but has yet to be realized in practice. Previous high-quality, color holographic displays have made either a 3$\times$ sacrifice on frame rate by using a sequential color illumination scheme or used more than one spatial light modulator (SLM) and/or bulky, complex optical setups. The reduced frame rate of sequential color introduces distracting judder and color fringing in the presence of head motion while the form factor of current simultaneous color systems is incompatible with a head-mounted display. In this work, we propose a framework for simultaneous color holography that allows the use of the full SLM frame rate while maintaining a compact and simple optical setup. Simultaneous color holograms are optimized through the use of a perceptual loss function, a physics-based neural network wavefront propagator, and a camera-calibrated forward model. We measurably improve hologram quality compared to other simultaneous color methods and move one step closer to the realization of color holographic displays for AR/VR.

physics.optics↗

arXiv2023

Multisource Holography

Holographic displays promise several benefits including high quality 3D imagery, accurate accommodation cues, and compact form-factors. However, holography relies on coherent illumination which can create undesirable speckle noise in the final image. Although smooth phase holograms can be speckle-free, their non-uniform eyebox makes them impractical, and speckle mitigation with partially coherent sources also reduces resolution. Averaging sequential frames for speckle reduction requires high speed modulators and consumes temporal bandwidth that may be needed elsewhere in the system. In this work, we propose multisource holography, a novel architecture that uses an array of sources to suppress speckle in a single frame without sacrificing resolution. By using two spatial light modulators, arranged sequentially, each source in the array can be controlled almost independently to create a version of the target content with different speckle. Speckle is then suppressed when the contributions from the multiple sources are averaged at the image plane. We introduce an algorithm to calculate multisource holograms, analyze the design space, and demonstrate up to a 10 dB increase in peak signal-to-noise ratio compared to an equivalent single source system. Finally, we validate the concept with a benchtop experimental prototype by producing both 2D images and focal stacks with natural defocus cues.

cs.GR↗

arXiv2023

Stochastic Light Field Holography

The Visual Turing Test is the ultimate goal to evaluate the realism of holographic displays. Previous studies have focused on addressing challenges such as limited étendue and image quality over a large focal volume, but they have not investigated the effect of pupil sampling on the viewing experience in full 3D holograms. In this work, we tackle this problem with a novel hologram generation algorithm motivated by matching the projection operators of incoherent Light Field and coherent Wigner Function light transport. To this end, we supervise hologram computation using synthesized photographs, which are rendered on-the-fly using Light Field refocusing from stochastically sampled pupil states during optimization. The proposed method produces holograms with correct parallax and focus cues, which are important for passing the Visual Turing Test. We validate that our approach compares favorably to state-of-the-art CGH algorithms that use Light Field and Focal Stack supervision. Our experiments demonstrate that our algorithm significantly improves the realism of the viewing experience for a variety of different pupil states.

cs.CV↗

arXiv2022

Pupil-aware Holography

Holographic displays promise to deliver unprecedented display capabilities in augmented reality applications, featuring a wide field of view, wide color gamut, spatial resolution, and depth cues all in a compact form factor. While emerging holographic display approaches have been successful in achieving large etendue and high image quality as seen by a camera, the large etendue also reveals a problem that makes existing displays impractical: the sampling of the holographic field by the eye pupil. Existing methods have not investigated this issue due to the lack of displays with large enough etendue, and, as such, they suffer from severe artifacts with varying eye pupil size and location. We show that the holographic field as sampled by the eye pupil is highly varying for existing display setups, and we propose pupil-aware holography that maximizes the perceptual image quality irrespective of the size, location, and orientation of the eye pupil in a near-eye holographic display. We validate the proposed approach both in simulations and on a prototype holographic display and show that our method eliminates severe artifacts and significantly outperforms existing approaches.

cs.GR↗

arXiv2022

medXGAN: Visual Explanations for Medical Classifiers through a Generative Latent Space

Despite the surge of deep learning in the past decade, some users are skeptical to deploy these models in practice due to their black-box nature. Specifically, in the medical space where there are severe potential repercussions, we need to develop methods to gain confidence in the models' decisions. To this end, we propose a novel medical imaging generative adversarial framework, medXGAN (medical eXplanation GAN), to visually explain what a medical classifier focuses on in its binary predictions. By encoding domain knowledge of medical images, we are able to disentangle anatomical structure and pathology, leading to fine-grained visualization through latent interpolation. Furthermore, we optimize the latent space such that interpolation explains how the features contribute to the classifier's output. Our method outperforms baselines such as Gradient-Weighted Class Activation Mapping (Grad-CAM) and Integrated Gradients in localization and explanatory ability. Additionally, a combination of the medXGAN with Integrated Gradients can yield explanations more robust to noise. The code is available at: https://avdravid.github.io/medXGAN_page/.

cs.CV↗

arXiv2022

Investigating the Potential of Auxiliary-Classifier GANs for Image Classification in Low Data Regimes

Generative Adversarial Networks (GANs) have shown promise in augmenting datasets and boosting convolutional neural networks' (CNN) performance on image classification tasks. But they introduce more hyperparameters to tune as well as the need for additional time and computational power to train supplementary to the CNN. In this work, we examine the potential for Auxiliary-Classifier GANs (AC-GANs) as a 'one-stop-shop' architecture for image classification, particularly in low data regimes. Additionally, we explore modifications to the typical AC-GAN framework, changing the generator's latent space sampling scheme and employing a Wasserstein loss with gradient penalty to stabilize the simultaneous training of image synthesis and classification. Through experiments on images of varying resolutions and complexity, we demonstrate that AC-GANs show promise in image classification, achieving competitive performance with standard CNNs. These methods can be employed as an 'all-in-one' framework with particular utility in the absence of large amounts of training data.

cs.CV↗

arXiv2021

Anatomy of a superorganism -- structure and growth dynamics of army ant bivouacs

Beyond unicellular and multicellular organisms, there is a third type of structural complexity in living animals: that of the mechanical self-assembly of groups of distinct multicellular organisms into dynamical, functional structures. One of the most striking examples of such structures is the army ant bivouac, a nest which self-assembles solely from the interconnected bodies of hundreds of thousands of individuals. These bivouacs are difficult to study because they rapidly disassemble when disturbed, and hence little is known about the structure and rules that individuals follow during their formation. Here we use a custom-built Computed Tomography scanner to investigate the details of the internal structure and growth process of army ant bivouacs. We show that bivouacs are heterogeneous structures, which throughout their growth maintain a thick shell surrounding a less dense interior that contains empty spaces akin to nest chambers. We find that ants within the bivouac do not carry more than approximately eight times their weight regardless of the size of the structure or their position within it. This observation suggests that bivouac size is not limited by physical constraints of the ants' morphology. This study brings us closer to understanding the rules used by individuals to govern the formation of these exceptional superorganismal structures, and provides insight into how to create engineered self-assembling systems with, for instance, swarms of robots or active matter.

q-bio.QM↗

arXiv2021

Improving Acquisition Speed of X-Ray Ptychography through Spatial Undersampling and Regularization

X-ray ptychography is one of the versatile techniques for nanometer resolution imaging. The magnitude of the diffraction patterns is recorded on a detector and the phase of the diffraction patterns is estimated using phase retrieval techniques. Most phase retrieval algorithms make the solution well-posed by relying on the constraints imposed by the overlapping region between neighboring diffraction pattern samples. As the overlap between neighboring diffraction patterns reduces, the problem becomes ill-posed and the object cannot be recovered. To avoid the ill-posedness, we investigate the effect of regularizing the phase retrieval algorithm with image priors for various overlap ratios between the neighboring diffraction patterns. We show that the object can be faithfully reconstructed at low overlap ratios by regularizing the phase retrieval algorithm with image priors such as Total-Variation and Structure Tensor Prior. We also show the effectiveness of our proposed algorithm on real data acquired from an IC chip with a coherent X-ray beam.

eess.IV↗

arXiv2021

SkinScan: Low-Cost 3D-Scanning for Dermatologic Diagnosis and Documentation

The utilization of computational photography becomes increasingly essential in the medical field. Today, imaging techniques for dermatology range from two-dimensional (2D) color imagery with a mobile device to professional clinical imaging systems measuring additional detailed three-dimensional (3D) data. The latter are commonly expensive and not accessible to a broad audience. In this work, we propose a novel system and software framework that relies only on low-cost (and even mobile) commodity devices present in every household to measure detailed 3D information of the human skin with a 3D-gradient-illumination-based method. We believe that our system has great potential for early-stage diagnosis and monitoring of skin diseases, especially in vastly populated or underdeveloped areas.

eess.IV↗

arXiv2020

2-Step Sparse-View CT Reconstruction with a Domain-Specific Perceptual Network

Computed tomography is widely used to examine internal structures in a non-destructive manner. To obtain high-quality reconstructions, one typically has to acquire a densely sampled trajectory to avoid angular undersampling. However, many scenarios require a sparse-view measurement leading to streak-artifacts if unaccounted for. Current methods do not make full use of the domain-specific information, and hence fail to provide reliable reconstructions for highly undersampled data. We present a novel framework for sparse-view tomography by decoupling the reconstruction into two steps: First, we overcome its ill-posedness using a super-resolution network, SIN, trained on the sparse projections. The intermediate result allows for a closed-form tomographic reconstruction with preserved details and highly reduced streak-artifacts. Second, a refinement network, PRN, trained on the reconstructions reduces any remaining artifacts. We further propose a light-weight variant of the perceptual-loss that enhances domain-specific information, boosting restoration accuracy. Our experiments demonstrate an improvement over current solutions by 4 dB.

eess.IV↗