Sven Lončarić

Ivan Sabolić, Marin Oršić, Josip Šarić et al.

Supervised fine-tuning is the predominant approach for adapting autoregressive vision-language models to downstream tasks. Recent work has shown that this paradigm is highly vulnerable to backdoor attacks, and that existing defenses are ineffective in open-ended generation settings. In response, we propose BYORn, a backdoor-robust fine-tuning framework motivated by the observation that poisoned target responses are often semantically implausible given the corresponding image-text inputs and a pretrained model. BYORn identifies such misaligned responses and dynamically replaces them with alternative responses generated by the model, thereby breaking the correlation between triggers and target outputs. The resulting objective gradient corresponds to the gradient of the empirical estimate of the population risk upper bound over the clean data distribution. Empirically, BYORn consistently improves robustness to backdoor attacks while preserving clean-task performance, establishing a new trade-off frontier between generalization and attack success rate. Finally, we demonstrate that BYORn remains effective against adaptive attacks specifically designed to circumvent the proposed defense.

Lovre Antonio Budimir, Yushi Guan, Steve Ryhner et al.

3D Language Gaussian Splatting (3DLGS) augments 3D Gaussian Splatting with language-aligned visual features for open-vocabulary 3D scene understanding. A core challenge is efficiently associating high-dimensional vision-language embeddings with millions of 3D Gaussians while preserving efficient feature rendering for text-based querying. Existing methods either store dense features directly on Gaussians, causing high storage costs and slow rendering, or learn compact representations through expensive per-scene optimization with repeated feature rasterization. No existing method simultaneously achieves fast 3D semantic reconstruction, efficient storage, and fast rendering. We propose SCOUP (Sparse COde UPlifting), which addresses all three by decoupling language representation learning from 3D Gaussian optimization. Rather than working directly in 3D, we learn sparse codebook-based representations entirely using features associated with 2D image regions, associating each region with a sparse set of codebook coefficients. We then uplift these coefficients to 3D Gaussians with our weighted sparse aggregation using Gaussian-to-pixel associations, where each Gaussian accumulates coefficients over codebook atoms across views. Top-$K$ filtering then extracts the most dominant multi-view coefficients per Gaussian, enabling efficient storage and fast rendering. Our method achieves up to $400\times$ training speedup while being $3\times$ more memory efficient during training compared to the state-of-the-art in rendering speed. Across multiple benchmarks, SCOUP matches or outperforms existing methods in open-vocabulary querying accuracy.

Egor Ershov, Alex Savchik, Illya Semenkov et al.

Computational color constancy has the important task of reducing the influence of the scene illumination on the object colors. As such, it is an essential part of the image processing pipelines of most digital cameras. One of the important parts of the computational color constancy is illumination estimation, i.e. estimating the illumination color. When an illumination estimation method is proposed, its accuracy is usually reported by providing the values of error metrics obtained on the images of publicly available datasets. However, over time it has been shown that many of these datasets have problems such as too few images, inappropriate image quality, lack of scene diversity, absence of version tracking, violation of various assumptions, GDPR regulation violation, lack of additional shooting procedure info, etc. In this paper, a new illumination estimation dataset is proposed that aims to alleviate many of the mentioned problems and to help the illumination estimation research. It consists of 4890 images with known illumination colors as well as with additional semantic data that can further make the learning process more accurate. Due to the usage of the SpyderCube color target, for every image there are two ground-truth illumination records covering different directions. Because of that, the dataset can be used for training and testing of methods that perform single or two-illuminant estimation. This makes it superior to many similar existing datasets. The datasets, it's smaller version SimpleCube++, and the accompanying code are available at https://github.com/Visillect/CubePlusPlus/.

Nikola Banić, Karlo Koščević, Marko Subašić et al.

Implementing color constancy as a pre-processing step in contemporary digital cameras is of significant importance as it removes the influence of scene illumination on object colors. Several benchmark color constancy datasets have been created for the purpose of developing and testing new color constancy methods. However, they all have numerous drawbacks including a small number of images, erroneously extracted ground-truth illuminations, long histories of misuses, violations of their stated assumptions, etc. To overcome such and similar problems, in this paper a color constancy benchmark dataset generator is proposed. For a given camera sensor it enables generation of any number of realistic raw images taken in a subset of the real world, namely images of printed photographs. Datasets with such images share many positive features with other existing real-world datasets, while some of the negative features are completely eliminated. The generated images can be successfully used to train methods that afterward achieve high accuracy on real-world datasets. This opens the way for creating large enough datasets for advanced deep learning techniques. Experimental results are presented and discussed. The source code is available at http://www.fer.unizg.hr/ipg/resources/color_constancy/.

Nikola Banić, Sven Lončarić

In the image processing pipeline of almost every digital camera there is a part dedicated to computational color constancy i.e. to removing the influence of illumination on the colors of the image scene. Some of the best known illumination estimation methods are the so called statistics-based methods. They are less accurate than the learning-based illumination estimation methods, but they are faster and simpler to implement in embedded systems, which is one of the reasons for their widespread usage. Although in the relevant literature it often appears as if they require no training, this is not true because they have parameter values that need to be fine-tuned in order to be more accurate. In this paper it is first shown that the accuracy of statistics-based methods reported in most papers was not obtained by means of the necessary cross-validation, but by using the whole benchmark datasets for both training and testing. After that the corrected results are given for the best known benchmark datasets. Finally, the so called green stability assumption is proposed that can be used to fine-tune the values of the parameters of the statistics-based methods by using only non-calibrated images without known ground-truth illumination. The obtained accuracy is practically the same as when using calibrated training images, but the whole process is much faster. The experimental results are presented and discussed. The source code is available at http://www.fer.unizg.hr/ipg/resources/color_constancy/.

Franko Šikić, Donik Vršnak, Sven Lončarić

In this paper, we present a survey of deep learning-based methods for the regression of gaze direction vector from head and eye images. We describe in detail numerous published methods with a focus on the input data, architecture of the model, and loss function used to supervise the model. Additionally, we present a list of datasets that can be used to train and evaluate gaze direction regression methods. Furthermore, we noticed that the results reported in the literature are often not comparable one to another due to differences in the validation or even test subsets used. To address this problem, we re-evaluated several methods on the commonly used in-the-wild Gaze360 dataset using the same validation setup. The experimental results show that the latest methods, although claiming state-of-the-art results, significantly underperform compared with some older methods. Finally, we show that the temporal models outperform the static models under static test conditions.

Franko Šikić, Sven Lončarić

Out-of-stock (OOS) detection is a very important retail verification process that aims to infer the unavailability of products in their designated areas on the shelf. In this paper, we introduce OOS-DSD, a novel deep learning-based method that advances OOS detection through auxiliary learning. In particular, we extend a well-established YOLOv8 object detection architecture with additional convolutional branches to simultaneously detect OOS, segment products, and estimate scene depth. While OOS detection and product segmentation branches are trained using ground truth data, the depth estimation branch is trained using pseudo-labeled annotations produced by the state-of-the-art (SOTA) depth estimation model Depth Anything V2. Furthermore, since the aforementioned pseudo-labeled depth estimates display relative depth, we propose an appropriate depth normalization procedure that stabilizes the training process. The experimental results show that the proposed method surpassed the performance of the SOTA OOS detection methods by 1.8% of the mean average precision (mAP). In addition, ablation studies confirm the effectiveness of auxiliary learning and the proposed depth normalization procedure, with the former increasing mAP by 3.7% and the latter by 4.2%.

Luka Posilović, Duje Medak, Marko Subasic et al.

Non-destructive testing is a set of techniques for defect detection in materials. While the set of imaging techniques are manifold, ultrasonic imaging is the one used the most. The analysis is mainly performed by human inspectors manually analyzing recorded images. The low number of defects in real ultrasonic inspections and legal issues considering data from such inspections make it difficult to obtain proper results from automatic ultrasonic image (B-scan) analysis. In this paper, we present a novel deep learning Generative Adversarial Network model for generating ultrasonic B-scans with defects in distinct locations. Furthermore, we show that generated B-scans can be used for synthetic data augmentation, and can improve the performance of deep convolutional neural object detection networks. Our novel method is demonstrated on a dataset of almost 4000 B-scans with more than 6000 annotated defects. Defect detection performance when training on real data yielded average precision of 71%. By training only on generated data the results increased to 72.1%, and by mixing generated and real data we achieve 75.7% average precision. We believe that synthetic data generation can generalize to other challenges with limited datasets and could be used for training human personnel.

Egor Ershov, Alex Savchik, Ilya Semenkov et al.

Illumination estimation is the essential step of computational color constancy, one of the core parts of various image processing pipelines of modern digital cameras. Having an accurate and reliable illumination estimation is important for reducing the illumination influence on the image colors. To motivate the generation of new ideas and the development of new algorithms in this field, the 2nd Illumination estimation challenge~(IEC\#2) was conducted. The main advantage of testing a method on a challenge over testing in on some of the known datasets is the fact that the ground-truth illuminations for the challenge test images are unknown up until the results have been submitted, which prevents any potential hyperparameter tuning that may be biased. The challenge had several tracks: general, indoor, and two-illuminant with each of them focusing on different parameters of the scenes. Other main features of it are a new large dataset of images (about 5000) taken with the same camera sensor model, a manual markup accompanying each image, diverse content with scenes taken in numerous countries under a huge variety of illuminations extracted by using the SpyderCube calibration object, and a contest-like markup for the images from the Cube+ dataset that was used in IEC\#1. This paper focuses on the description of the past two challenges, algorithms which won in each track, and the conclusions that were drawn based on the results obtained during the 1st and 2nd challenge that can be useful for similar future developments.

Julian Lo, Morgan Heisler, Vinicius Vanzan et al.

Purpose: Optical Coherence Tomography Angiography (OCT-A) permits visualization of the changes to the retinal circulation due to diabetic retinopathy (DR), a microvascular complication of diabetes. We demonstrate accurate segmentation of the vascular morphology for the superficial capillary plexus and deep vascular complex (SCP and DVC) using a convolutional neural network (CNN) for quantitative analysis. Methods: Retinal OCT-A with a 6x6mm field of view (FOV) were acquired using a Zeiss PlexElite. Multiple-volume acquisition and averaging enhanced the vessel network contrast used for training the CNN. We used transfer learning from a CNN trained on 76 images from smaller FOVs of the SCP acquired using different OCT systems. Quantitative analysis of perfusion was performed on the automated vessel segmentations in representative patients with DR. Results: The automated segmentations of the OCT-A images maintained the hierarchical branching and lobular morphologies of the SCP and DVC, respectively. The network segmented the SCP with an accuracy of 0.8599, and a Dice index of 0.8618. For the DVC, the accuracy was 0.7986, and the Dice index was 0.8139. The inter-rater comparisons for the SCP had an accuracy and Dice index of 0.8300 and 0.6700, respectively, and 0.6874 and 0.7416 for the DVC. Conclusions: Transfer learning reduces the amount of manually-annotated images required, while producing high quality automatic segmentations of the SCP and DVC. Using high quality training data preserves the characteristic appearance of the capillary networks in each layer. Translational Relevance: Accurate retinal microvasculature segmentation with the CNN results in improved perfusion analysis in diabetic retinopathy.

Morgan Heisler, Forson Chan, Zaid Mammo et al.

Automatic quantification of perifoveal vessel densities in optical coherence tomography angiography (OCT-A) images face challenges such as variable intra- and inter-image signal to noise ratios, projection artefacts from outer vasculature layers, and motion artefacts. This study demonstrates the utility of deep neural networks for automatic quantification of foveal avascular zone (FAZ) parameters and perifoveal vessel density of OCT-A images in healthy and diabetic eyes. OCT-A images of the foveal region were acquired using three OCT-A systems: a 1060nm Swept Source (SS)-OCT prototype, RTVue XR Avanti (Optovue Inc., Fremont, CA), and the ZEISS Angioplex (Carl Zeiss Meditec, Dublin, CA). Automated segmentation was then performed using a deep neural network. Four FAZ morphometric parameters (area, min/max diameter, and eccentricity) and perifoveal vessel density were used as outcome measures. The accuracy, sensitivity and specificity of the DNN vessel segmentations were comparable across all three device platforms. No significant difference between the means of the measurements from automated and manual segmentations were found for any of the outcome measures on any system. The intraclass correlation coefficient (ICC) was also good (> 0.51) for all measurements. Automated deep learning vessel segmentation of OCT-A may be suitable for both commercial and research purposes for better quantification of the retinal circulation.

Andrija Štajduhar, Tomislav Lipić, Goran Sedmak et al.

In this paper, we present a novel method for analysis and segmentation of laminar structure of the cortex based on tissue characteristics whose change across the gray matter underlies distinctive between cortical layers. We develop and analyze features of individual neurons to investigate changes in cytoarchitectonic differentiation and present a novel high-performance, automated framework for neuron-level histological image analysis. Local tissue and cell descriptors such as density, neuron size and other measures are used for development of more complex neuron features used in machine learning model trained on data manually labeled by three human experts. Final neuron layer classifications were obtained by training a separate model for each expert and combining their probability outputs. Importances of developed neuron features on both global model level and individual prediction level are presented and discussed.

Nikola Banić, Karlo Koš{č}ević, Marko Subašić et al.

The pipelines of digital cameras contain a part for computational color constancy, which aims to remove the influence of the illumination on the scene colors. One of the best known and most widely used benchmark datasets for this problem is the Color Checker dataset. However, due to the improper handling of the black level in its images, this dataset has been widely misused and while some recent publications tried to alleviate the problem, they nevertheless erred and created additional wrong data. This paper gives a history of the Color Checker dataset usage, it describes the origins and reasons for its misuses, and it explains the old and new mistakes introduced in the most recent publications that tried to handle the issue. This should, hopefully, help to prevent similar future misuses.

Andrija Štajduhar, Domagoj Džaja, Miloš Judaš et al.

In this paper, we present a novel use of an anisotropic diffusion model for automatic detection of neurons in histological sections of the adult human brain cortex. We use a partial differential equation model to process high resolution images to acquire locations of neuronal bodies. We also present a novel approach in model training and evaluation that considers variability among the human experts, addressing the issue of existence and correctness of the golden standard for neuron and cell counting, used in most of relevant papers. Our method, trained on dataset manually labeled by three experts, has correctly distinguished over 95% of neuron bodies in test data, doing so in time much shorter than other comparable methods.

Nikola Banić, Karlo Koščević, Sven Lončarić

Most digital camera pipelines use color constancy methods to reduce the influence of illumination and camera sensor on the colors of scene objects. The highest accuracy of color correction is obtained with learning-based color constancy methods, but they require a significant amount of calibrated training images with known ground-truth illumination. Such calibration is time consuming, preferably done for each sensor individually, and therefore a major bottleneck in acquiring high color constancy accuracy. Statistics-based methods do not require calibrated training images, but they are less accurate. In this paper an unsupervised learning-based method is proposed that learns its parameter values after approximating the unknown ground-truth illumination of the training images, thus avoiding calibration. In terms of accuracy the proposed method outperforms all statistics-based and many learning-based methods. An extension of the method is also proposed, which learns the needed parameters from non-calibrated images taken with one sensor and which can then be successfully applied to images taken with another sensor. This effectively enables inter-camera unsupervised learning for color constancy. Additionally, a new high quality color constancy benchmark dataset with 1707 calibrated images is created, used for testing, and made publicly available. The results are presented and discussed. The source code and the dataset are available at http://www.fer.unizg.hr/ipg/resources/color_constancy/.

Tomislav Petković, Sven Lončarić

The Hough transform is one of the most common methods for line detection. In this paper we propose a novel extension of the regular Hough transform. The proposed extension combines the extension of the accumulator space and the local gradient orientation resulting in clutter reduction and yielding more prominent peaks, thus enabling better line identification. We demonstrate benefits in applications such as visual quality inspection and rectangle detection.

Sven Lončarić, Siniša Šegvić

Proceedings of the Second Croatian Computer Vision Workshop (CCVW 2013, http://www.fer.unizg.hr/crv/ccvw2013) held September 19, 2013, in Zagreb, Croatia. Workshop was organized by the Center of Excellence for Computer Vision of the University of Zagreb.

Tomislav Petković, Darko Jurić, Sven Lončarić

Most visual quality inspections in discrete manufacturing are composed of length, surface, angle or intensity measurements. Those are implemented as end-user configurable inspection tools that should not require an image processing expert to set up. Currently available software solutions providing such capability use a flowchart based programming environment, but do not fully address an inspection flowchart robustness and can require a redefinition of the flowchart if a small variation is introduced. In this paper we propose an acquire-register-analyze image processing pattern designed for discrete manufacturing that aims to increase the robustness of the inspection flowchart by consistently addressing variations in product position, orientation and size. A proposed pattern is transparent to the end-user and simplifies the flowchart. We describe a developed software solution that is a practical implementation of the proposed pattern. We give an example of its real-life use in industrial production of electric components.