Rafael E. Rivadeneira, Angel Sappa, Boris X. Vintimilla, Jin Kim, Dogun Kim, Zhihao Li, et al. (2022). Thermal Image Super-Resolution Challenge Results – PBVS 2022. In IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (pp. 418–426).
Abstract: This paper presents results from the third Thermal Image Super-Resolution (TISR) challenge organized in the Perception Beyond the Visible Spectrum (PBVS) 2022 workshop. The challenge uses the same thermal image dataset as the first two challenges, with 951 training images and 50 validation images at each resolution. A set of 20 images was kept aside for testing. The evaluation tasks were to measure the PSNR and SSIM between the SR image and the ground truth (HR thermal noisy image downsampled by four), and also to measure the PSNR and SSIM between the SR image and the semi-registered HR image (acquired with another camera). The results outperformed those from last year’s challenge, improving both evaluation metrics. This year, almost 100 teams participants registered for the challenge, showing the community’s interest in this hot topic.
|
Rafael E. Rivadeneira, Angel Sappa, & Boris X. Vintimilla. (2022). Thermal Image Super-Resolution: A Novel Unsupervised Approach. In International Joint Conference on Computer Vision, Imaging and Computer Graphics (Vol. 1474, 495–506).
Abstract: This paper proposes the use of a CycleGAN architecture for thermal image super-resolution under a transfer domain strategy, where middle-resolution images from one camera are transferred to a higher resolution domain of another camera. The proposed approach is trained with a large dataset acquired using three thermal cameras at different resolutions. An unsupervised learning process is followed to train the architecture. Additional loss function is proposed trying to improve results from the state of the art approaches. Following the first thermal image super-resolution challenge (PBVS-CVPR2020) evaluations are performed. A comparison with previous works is presented showing the proposed approach reaches the best results.
|
Marc Oliu, Sarah Adel Bargal, Stan Sclaroff, Xavier Baro, & Sergio Escalera. (2022). Multi-varied Cumulative Alignment for Domain Adaptation. In 6th International Conference on Image Analysis and Processing (Vol. 13232, 324–334). LNCS.
Abstract: Domain Adaptation methods can be classified into two basic families of approaches: non-parametric and parametric. Non-parametric approaches depend on statistical indicators such as feature covariances to minimize the domain shift. Non-parametric approaches tend to be fast to compute and require no additional parameters, but they are unable to leverage probability density functions with complex internal structures. Parametric approaches, on the other hand, use models of the probability distributions as surrogates in minimizing the domain shift, but they require additional trainable parameters to model these distributions. In this work, we propose a new statistical approach to minimizing the domain shift based on stochastically projecting and evaluating the cumulative density function in both domains. As with non-parametric approaches, there are no additional trainable parameters. As with parametric approaches, the internal structure of both domains’ probability distributions is considered, thus leveraging a higher amount of information when reducing the domain shift. Evaluation on standard datasets used for Domain Adaptation shows better performance of the proposed model compared to non-parametric approaches while being competitive with parametric ones. (Code available at: https://github.com/moliusimon/mca).
Keywords: Domain Adaptation; Computer vision; Neural networks
|
Ajian Liu, Chenxu Zhao, Zitong Yu, Jun Wan, Anyang Su, Xing Liu, et al. (2022). Contrastive Context-Aware Learning for 3D High-Fidelity Mask Face Presentation Attack Detection. TIForensicSEC - IEEE Transactions on Information Forensics and Security, 17, 2497–2507.
Abstract: Face presentation attack detection (PAD) is essential to secure face recognition systems primarily from high-fidelity mask attacks. Most existing 3D mask PAD benchmarks suffer from several drawbacks: 1) a limited number of mask identities, types of sensors, and a total number of videos; 2) low-fidelity quality of facial masks. Basic deep models and remote photoplethysmography (rPPG) methods achieved acceptable performance on these benchmarks but still far from the needs of practical scenarios. To bridge the gap to real-world applications, we introduce a large-scale Hi gh- Fi delity Mask dataset, namely HiFiMask . Specifically, a total amount of 54,600 videos are recorded from 75 subjects with 225 realistic masks by 7 new kinds of sensors. Along with the dataset, we propose a novel C ontrastive C ontext-aware L earning (CCL) framework. CCL is a new training methodology for supervised PAD tasks, which is able to learn by leveraging rich contexts accurately (e.g., subjects, mask material and lighting) among pairs of live faces and high-fidelity mask attacks. Extensive experimental evaluations on HiFiMask and three additional 3D mask datasets demonstrate the effectiveness of our method. The codes and dataset will be released soon.
|
Hugo Bertiche, Meysam Madadi, & Sergio Escalera. (2022). Neural Cloth Simulation. ACMTGraph - ACM Transactions on Graphics, 41(6), 1–14.
Abstract: We present a general framework for the garment animation problem through unsupervised deep learning inspired in physically based simulation. Existing trends in the literature already explore this possibility. Nonetheless, these approaches do not handle cloth dynamics. Here, we propose the first methodology able to learn realistic cloth dynamics unsupervisedly, and henceforth, a general formulation for neural cloth simulation. The key to achieve this is to adapt an existing optimization scheme for motion from simulation based methodologies to deep learning. Then, analyzing the nature of the problem, we devise an architecture able to automatically disentangle static and dynamic cloth subspaces by design. We will show how this improves model performance. Additionally, this opens the possibility of a novel motion augmentation technique that greatly improves generalization. Finally, we show it also allows to control the level of motion in the predictions. This is a useful, never seen before, tool for artists. We provide of detailed analysis of the problem to establish the bases of neural cloth simulation and guide future research into the specifics of this domain.
ACM Transactions on GraphicsVolume 41Issue 6December 2022 Article No.: 220pp 1–
|
Joakim Bruslund Haurum, Meysam Madadi, Sergio Escalera, & Thomas B. Moeslund. (2022). Multi-scale hybrid vision transformer and Sinkhorn tokenizer for sewer defect classification. AC - Automation in Construction, 144, 104614.
Abstract: A crucial part of image classification consists of capturing non-local spatial semantics of image content. This paper describes the multi-scale hybrid vision transformer (MSHViT), an extension of the classical convolutional neural network (CNN) backbone, for multi-label sewer defect classification. To better model spatial semantics in the images, features are aggregated at different scales non-locally through the use of a lightweight vision transformer, and a smaller set of tokens was produced through a novel Sinkhorn clustering-based tokenizer using distinct cluster centers. The proposed MSHViT and Sinkhorn tokenizer were evaluated on the Sewer-ML multi-label sewer defect classification dataset, showing consistent performance improvements of up to 2.53 percentage points.
Keywords: Sewer Defect Classification; Vision Transformers; Sinkhorn-Knopp; Convolutional Neural Networks; Closed-Circuit Television; Sewer Inspection
|
Aitor Alvarez-Gila, Joost Van de Weijer, Yaxing Wang, & Estibaliz Garrote. (2022). MVMO: A Multi-Object Dataset for Wide Baseline Multi-View Semantic Segmentation. In 29th IEEE International Conference on Image Processing.
Abstract: We present MVMO (Multi-View, Multi-Object dataset): a synthetic dataset of 116,000 scenes containing randomly placed objects of 10 distinct classes and captured from 25 camera locations in the upper hemisphere. MVMO comprises photorealistic, path-traced image renders, together with semantic segmentation ground truth for every view. Unlike existing multi-view datasets, MVMO features wide baselines between cameras and high density of objects, which lead to large disparities, heavy occlusions and view-dependent object appearance. Single view semantic segmentation is hindered by self and inter-object occlusions that could benefit from additional viewpoints. Therefore, we expect that MVMO will propel research in multi-view semantic segmentation and cross-view semantic transfer. We also provide baselines that show that new research is needed in such fields to exploit the complementary information of multi-view setups 1 .
Keywords: multi-view; cross-view; semantic segmentation; synthetic dataset
|
Nil Ballus, Bhalaji Nagarajan, & Petia Radeva. (2022). Opt-SSL: An Enhanced Self-Supervised Framework for Food Recognition. In 10th Iberian Conference on Pattern Recognition and Image Analysis (Vol. 13256). LNCS.
Abstract: Self-supervised Learning has been showing upbeat performance in several computer vision tasks. The popular contrastive methods make use of a Siamese architecture with different loss functions. In this work, we go deeper into two very recent state of the art frameworks, namely, SimSiam and Barlow Twins. Inspired by them, we propose a new self-supervised learning method we call Opt-SSL that combines both image and feature contrasting. We validate the proposed method on the food recognition task, showing that our proposed framework enables the self-learning networks to learn better visual representations.
Keywords: Self-supervised; Contrastive learning; Food recognition
|
Sergi Garcia Bordils, George Tom, Sangeeth Reddy, Minesh Mathew, Marçal Rusiñol, C.V. Jawahar, et al. (2022). Read While You Drive-Multilingual Text Tracking on the Road. In 15th IAPR International workshop on document analysis systems (Vol. 13237, 756–770). LNCS.
Abstract: Visual data obtained during driving scenarios usually contain large amounts of text that conveys semantic information necessary to analyse the urban environment and is integral to the traffic control plan. Yet, research on autonomous driving or driver assistance systems typically ignores this information. To advance research in this direction, we present RoadText-3K, a large driving video dataset with fully annotated text. RoadText-3K is three times bigger than its predecessor and contains data from varied geographical locations, unconstrained driving conditions and multiple languages and scripts. We offer a comprehensive analysis of tracking by detection and detection by tracking methods exploring the limits of state-of-the-art text detection. Finally, we propose a new end-to-end trainable tracking model that yields state-of-the-art results on this challenging dataset. Our experiments demonstrate the complexity and variability of RoadText-3K and establish a new, realistic benchmark for scene text tracking in the wild.
|
Francesco Pelosin, Saurav Jha, Andrea Torsello, Bogdan Raducanu, & Joost Van de Weijer. (2022). Towards exemplar-free continual learning in vision transformers: an account of attention, functional and weight regularization. In IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).
Abstract: In this paper, we investigate the continual learning of Vision Transformers (ViT) for the challenging exemplar-free scenario, with special focus on how to efficiently distill the knowledge of its crucial self-attention mechanism (SAM). Our work takes an initial step towards a surgical investigation of SAM for designing coherent continual learning methods in ViTs. We first carry out an evaluation of established continual learning regularization techniques. We then examine the effect of regularization when applied to two key enablers of SAM: (a) the contextualized embedding layers, for their ability to capture well-scaled representations with respect to the values, and (b) the prescaled attention maps, for carrying value-independent global contextual information. We depict the perks of each distilling strategy on two image recognition benchmarks (CIFAR100 and ImageNet-32) – while (a) leads to a better overall accuracy, (b) helps enhance the rigidity by maintaining competitive performances. Furthermore, we identify the limitation imposed by the symmetric nature of regularization losses. To alleviate this, we propose an asymmetric variant and apply it to the pooled output distillation (POD) loss adapted for ViTs. Our experiments confirm that introducing asymmetry to POD boosts its plasticity while retaining stability across (a) and (b). Moreover, we acknowledge low forgetting measures for all the compared methods, indicating that ViTs might be naturally inclined continual learners. 1
Keywords: Learning systems; Weight measurement; Image recognition; Surgery; Benchmark testing; Transformers; Stability analysis
|
Hector Laria Mantecon, Yaxing Wang, Joost Van de Weijer, & Bogdan Raducanu. (2022). Transferring Unconditional to Conditional GANs With Hyper-Modulation. In IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).
Abstract: GANs have matured in recent years and are able to generate high-resolution, realistic images. However, the computational resources and the data required for the training of high-quality GANs are enormous, and the study of transfer learning of these models is therefore an urgent topic. Many of the available high-quality pretrained GANs are unconditional (like StyleGAN). For many applications, however, conditional GANs are preferable, because they provide more control over the generation process, despite often suffering more training difficulties. Therefore, in this paper, we focus on transferring from high-quality pretrained unconditional GANs to conditional GANs. This requires architectural adaptation of the pretrained GAN to perform the conditioning. To this end, we propose hyper-modulated generative networks that allow for shared and complementary supervision. To prevent the additional weights of the hypernetwork to overfit, with subsequent mode collapse on small target domains, we introduce a self-initialization procedure that does not require any real data to initialize the hypernetwork parameters. To further improve the sample efficiency of the transfer, we apply contrastive learning in the discriminator, which effectively works on very limited batch sizes. In extensive experiments, we validate the efficiency of the hypernetworks, self-initialization and contrastive loss for knowledge transfer on standard benchmarks.
|
Vacit Oguz Yazici, Joost Van de Weijer, & Longlong Yu. (2022). Visual Transformers with Primal Object Queries for Multi-Label Image Classification. In 26th International Conference on Pattern Recognition.
Abstract: Multi-label image classification is about predicting a set of class labels that can be considered as orderless sequential data. Transformers process the sequential data as a whole, therefore they are inherently good at set prediction. The first vision-based transformer model, which was proposed for the object detection task introduced the concept of object queries. Object queries are learnable positional encodings that are used by attention modules in decoder layers to decode the object classes or bounding boxes using the region of interests in an image. However, inputting the same set of object queries to different decoder layers hinders the training: it results in lower performance and delays convergence. In this paper, we propose the usage of primal object queries that are only provided at the start of the transformer decoder stack. In addition, we improve the mixup technique proposed for multi-label classification. The proposed transformer model with primal object queries improves the state-of-the-art class wise F1 metric by 2.1% and 1.8%; and speeds up the convergence by 79.0% and 38.6% on MS-COCO and NUS-WIDE datasets respectively.
|
Swathikiran Sudhakaran, Sergio Escalera, & Oswald Lanz. (2023). Gate-Shift-Fuse for Video Action Recognition. TPAMI - IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(9), 10913–10928.
Abstract: Convolutional Neural Networks are the de facto models for image recognition. However 3D CNNs, the straight forward extension of 2D CNNs for video recognition, have not achieved the same success on standard action recognition benchmarks. One of the main reasons for this reduced performance of 3D CNNs is the increased computational complexity requiring large scale annotated datasets to train them in scale. 3D kernel factorization approaches have been proposed to reduce the complexity of 3D CNNs. Existing kernel factorization approaches follow hand-designed and hard-wired techniques. In this paper we propose Gate-Shift-Fuse (GSF), a novel spatio-temporal feature extraction module which controls interactions in spatio-temporal decomposition and learns to adaptively route features through time and combine them in a data dependent manner. GSF leverages grouped spatial gating to decompose input tensor and channel weighting to fuse the decomposed tensors. GSF can be inserted into existing 2D CNNs to convert them into an efficient and high performing spatio-temporal feature extractor, with negligible parameter and compute overhead. We perform an extensive analysis of GSF using two popular 2D CNN families and achieve state-of-the-art or competitive performance on five standard action recognition benchmarks.
Keywords: Action Recognition; Video Classification; Spatial Gating; Channel Fusion
|
Ayan Banerjee, Palaiahnakote Shivakumara, Parikshit Acharya, Umapada Pal, & Josep Llados. (2022). TWD: A New Deep E2E Model for Text Watermark Detection in Video Images. In 26th International Conference on Pattern Recognition.
Abstract: Text watermark detection in video images is challenging because text watermark characteristics are different from caption and scene texts in the video images. Developing a successful model for detecting text watermark, caption, and scene texts is an open challenge. This study aims at developing a new Deep End-to-End model for Text Watermark Detection (TWD), caption and scene text in video images. To standardize non-uniform contrast, quality, and resolution, we explore the U-Net3+ model for enhancing poor quality text without affecting high-quality text. Similarly, to address the challenges of arbitrary orientation, text shapes and complex background, we explore Stacked Hourglass Encoded Fourier Contour Embedding Network (SFCENet) by feeding the output of the U-Net3+ model as input. Furthermore, the proposed work integrates enhancement and detection models as an end-to-end model for detecting multi-type text in video images. To validate the proposed model, we create our own dataset (named TW-866), which provides video images containing text watermark, caption (subtitles), as well as scene text. The proposed model is also evaluated on standard natural scene text detection datasets, namely, ICDAR 2019 MLT, CTW1500, Total-Text, and DAST1500. The results show that the proposed method outperforms the existing methods. This is the first work on text watermark detection in video images to the best of our knowledge
Keywords: Deep learning; U-Net; FCENet; Scene text detection; Video text detection; Watermark text detection
|
Ahmed M. A. Salih, Ilaria Boscolo Galazzo, Federica Cruciani, Lorenza Brusini, & Petia Radeva. (2022). Investigating Explainable Artificial Intelligence for MRI-based Classification of Dementia: a New Stability Criterion for Explainable Methods. In 29th IEEE International Conference on Image Processing.
Abstract: Individuals diagnosed with Mild Cognitive Impairment (MCI) have shown an increased risk of developing Alzheimer’s Disease (AD). As such, early identification of dementia represents a key prognostic element, though hampered by complex disease patterns. Increasing efforts have focused on Machine Learning (ML) to build accurate classification models relying on a multitude of clinical/imaging variables. However, ML itself does not provide sensible explanations related to the model mechanism and feature contribution. Explainable Artificial Intelligence (XAI) represents the enabling technology in this framework, allowing to understand ML outcomes and derive human-understandable explanations. In this study, we aimed at exploring ML combined with MRI-based features and XAI to solve this classification problem and interpret the outcome. In particular, we propose a new method to assess the robustness of feature rankings provided by XAI methods, especially when multicollinearity exists. Our findings indicate that our method was able to disentangle the list of the informative features underlying dementia, with important implications for aiding personalized monitoring plans.
Keywords: Image processing; Stability criteria; Machine learning; Robustness; Alzheimer's disease; Monitoring
|
