Vassilis Athitsos

Mohammad Rezaei, Razieh Rastgoo, Vassilis Athitsos

3D hand pose estimation methods have made significant progress recently. However, the estimation accuracy is often far from sufficient for specific real-world applications, and thus there is significant room for improvement. This paper proposes TriHorn-Net, a novel model that uses specific innovations to improve hand pose estimation accuracy on depth images. The first innovation is the decomposition of the 3D hand pose estimation into the estimation of 2D joint locations in the depth image space (UV), and the estimation of their corresponding depths aided by two complementary attention maps. This decomposition prevents depth estimation, which is a more difficult task, from interfering with the UV estimations at both the prediction and feature levels. The second innovation is PixDropout, which is, to the best of our knowledge, the first appearance-based data augmentation method for hand depth images. Experimental results demonstrate that the proposed model outperforms the state-of-the-art methods on three public benchmark datasets. Our implementation is available at https://github.com/mrezaei92/TriHorn-Net.

Mohammad Rezaei, Farnaz Farahanipad, Alex Dillhoff et al.

Despite the significant progress that depth-based 3D hand pose estimation methods have made in recent years, they still require a large amount of labeled training data to achieve high accuracy. However, collecting such data is both costly and time-consuming. To tackle this issue, we propose a semi-supervised method to significantly reduce the dependence on labeled training data. The proposed method consists of two identical networks trained jointly: a teacher network and a student network. The teacher network is trained using both the available labeled and unlabeled samples. It leverages the unlabeled samples via a loss formulation that encourages estimation equivariance under a set of affine transformations. The student network is trained using the unlabeled samples with their pseudo-labels provided by the teacher network. For inference at test time, only the student network is used. Extensive experiments demonstrate that the proposed method outperforms the state-of-the-art semi-supervised methods by large margins.

Lana Do, Shasta Ihorn, Charity M. Pitcher-Cooper et al.

Audio description (AD) narrates visual elements in video for blind and low-vision audiences. Recent work has shown that giving novice describers an AI-generated draft to start from helps produce higher-quality AD and lowers the barrier to entry. What remains an open question is how draft quality shapes the editing process. We investigate this through GenAD, an AD generation pipeline that incorporates accessibility guidelines and contextual video information, and RefineAD, an editing interface for human revisions. Human-AI contributions are measured across text, timing, and delivery. In a within-subjects study, we compared authoring from scratch against editing AI drafts of varying quality. GenAD drafts cut completion time by more than half and significantly reduced cognitive load. In contrast, baseline drafts generated from simple, unguided prompts offered only modest benefits, pointing to a minimum quality threshold for effectiveness. Qualitative findings suggest this threshold is content-dependent; as visual complexity increases, so does the quality needed from AI drafts. We propose this as a design principle: effective AI assistance should clear a quality threshold suited to the target content, rather than simply be present.

Lana Do, Gio Jung, Juvenal Francisco Barajas et al.

Digital video is central to communication, education, and entertainment, but without audio description (AD), blind and low-vision audiences are excluded. While crowdsourced platforms and vision-language-models (VLMs) expand AD production, quality is rarely checked systematically. Existing evaluations rely on NLP metrics and short-clip guidelines, leaving questions about what constitutes quality for full-length content and how to assess it at scale. To address these questions, we first developed a multi-dimensional assessment framework for uninterrupted, full-length video, grounded in professional guidelines and refined by accessibility specialists. Second, we integrated this framework into a comprehensive methodological workflow, utilizing Item Response Theory, to assess the proficiency of VLM and human raters against expert-established ground truth. Findings suggest that while VLMs can approximate ground-truth ratings with high alignment, their reasoning was found to be less reliable and actionable than that of human respondents. These insights show the potential of hybrid evaluation systems that leverage VLMs alongside human oversight, offering a path towards scalable AD quality control.

Razieh Rastgoo, Kourosh Kiani, Sergio Escalera et al.

Sign Language is the dominant form of communication language used in the deaf and hearing-impaired community. To make an easy and mutual communication between the hearing-impaired and the hearing communities, building a robust system capable of translating the spoken language into sign language and vice versa is fundamental. To this end, sign language recognition and production are two necessary parts for making such a two-way system. Sign language recognition and production need to cope with some critical challenges. In this survey, we review recent advances in Sign Language Production (SLP) and related areas using deep learning. To have more realistic perspectives to sign language, we present an introduction to the Deaf culture, Deaf centers, psychological perspective of sign language, the main differences between spoken language and sign language. Furthermore, we present the fundamental components of a bi-directional sign language translation system, discussing the main challenges in this area. Also, the backbone architectures and methods in SLP are briefly introduced and the proposed taxonomy on SLP is presented. Finally, a general framework for SLP and performance evaluation, and also a discussion on the recent developments, advantages, and limitations in SLP, commenting on possible lines for future research are presented.

Zahra Anvari, Vassilis Athitsos

Digitized documents such as scientific articles, tax forms, invoices, contract papers, historic texts are widely used nowadays. These document images could be degraded or damaged due to various reasons including poor lighting conditions, shadow, distortions like noise and blur, aging, ink stain, bleed-through, watermark, stamp, etc. Document image enhancement plays a crucial role as a pre-processing step in many automated document analysis and recognition tasks such as character recognition. With recent advances in deep learning, many methods are proposed to enhance the quality of these document images. In this paper, we review deep learning-based methods, datasets, and metrics for six main document image enhancement tasks, including binarization, debluring, denoising, defading, watermark removal, and shadow removal. We summarize the recent works for each task and discuss their features, challenges, and limitations. We introduce multiple document image enhancement tasks that have received little to no attention, including over and under exposure correction, super resolution, and bleed-through removal. We identify several promising research directions and opportunities for future research.

Saif Sayed, Vassilis Athitsos

While many action recognition techniques have great success on public benchmarks, such performance is not necessarily replicated in real-world scenarios, where the data comes from specific application requirements. The specific real-world application that we are focusing on in this paper is cognitive assessment in children using cognitively demanding physical tasks. We created a system called Cross-Your-Body and recorded data, which is unique in several aspects, including the fact that the tasks have been designed by psychologists, the subjects are children, and the videos capture real-world usage, as they record children performing tasks during real-world assessment by psychologists. Other distinguishing features of our system is that it's scores can directly be translated to measure executive functioning which is one of the key factor to distinguish onset of ADHD in adolescent kids. Due to imprecise execution of actions performed by children, and the presence of fine-grained motion patterns, we systematically investigate and evaluate relevant methods on the recorded data. It is our goal that this system will be useful in advancing research in cognitive assessment of kids.

Reza Ghoddoosian, Saif Sayed, Vassilis Athitsos

This paper focuses on task recognition and action segmentation in weakly-labeled instructional videos, where only the ordered sequence of video-level actions is available during training. We propose a two-stream framework, which exploits semantic and temporal hierarchies to recognize top-level tasks in instructional videos. Further, we present a novel top-down weakly-supervised action segmentation approach, where the predicted task is used to constrain the inference of fine-grained action sequences. Experimental results on the popular Breakfast and Cooking 2 datasets show that our two-stream hierarchical task modeling significantly outperforms existing methods in top-level task recognition for all datasets and metrics. Additionally, using our task recognition framework in the proposed top-down action segmentation approach consistently improves the state of the art, while also reducing segmentation inference time by 80-90 percent.

Shiba Kuanar, Vassilis Athitsos, Dwarikanath Mahapatra et al.

Clear cell renal cell carcinoma (ccRCC) is one of the most common forms of intratumoral heterogeneity in the study of renal cancer. ccRCC originates from the epithelial lining of proximal convoluted renal tubules. These cells undergo abnormal mutations in the presence of Ki67 protein and create a lump-like structure through cell proliferation. Manual counting of tumor cells in the tissue-affected sections is one of the strongest prognostic markers for renal cancer. However, this procedure is time-consuming and also prone to subjectivity. These assessments are based on the physical cell appearance and suffer wide intra-observer variations. Therefore, better cell nucleus detection and counting techniques can be an important biomarker for the assessment of tumor cell proliferation in routine pathological investigations. In this paper, we introduce a deep learning-based detection model for cell classification on IHC stained histology images. These images are classified into binary classes to find the presence of Ki67 protein in cancer-affected nucleus regions. Our model maps the multi-scale pyramid features and saliency information from local bounded regions and predicts the bounding box coordinates through regression. Our method validates the impact of Ki67 expression across a cohort of four hundred histology images treated with localized ccRCC and compares our results with the existing state-of-the-art nucleus detection methods. The precision and recall scores of the proposed method are computed and compared on the clinical data sets. The experimental results demonstrate that our model improves the F1 score up to 86.3% and an average area under the Precision-Recall curve as 85.73%.

Reza Ghoddoosian, Saif Sayed, Vassilis Athitsos

This paper focuses on weakly-supervised action alignment, where only the ordered sequence of video-level actions is available for training. We propose a novel Duration Network, which captures a short temporal window of the video and learns to predict the remaining duration of a given action at any point in time with a level of granularity based on the type of that action. Further, we introduce a Segment-Level Beam Search to obtain the best alignment, that maximizes our posterior probability. Segment-Level Beam Search efficiently aligns actions by considering only a selected set of frames that have more confident predictions. The experimental results show that our alignments for long videos are more robust than existing models. Moreover, the proposed method achieves state of the art results in certain cases on the popular Breakfast and Hollywood Extended datasets.

Ashiq Imran, Vassilis Athitsos

Fine-Grained Visual Categorization (FGVC) is a challenging topic in computer vision. It is a problem characterized by large intra-class differences and subtle inter-class differences. In this paper, we tackle this problem in a weakly supervised manner, where neural network models are getting fed with additional data using a data augmentation technique through a visual attention mechanism. We perform domain adaptive knowledge transfer via fine-tuning on our base network model. We perform our experiment on six challenging and commonly used FGVC datasets, and we show competitive improvement on accuracies by using attention-aware data augmentation techniques with features derived from deep learning model InceptionV3, pre-trained on large scale datasets. Our method outperforms competitor methods on multiple FGVC datasets and showed competitive results on other datasets. Experimental studies show that transfer learning from large scale datasets can be utilized effectively with visual attention based data augmentation, which can obtain state-of-the-art results on several FGVC datasets. We present a comprehensive analysis of our experiments. Our method achieves state-of-the-art results in multiple fine-grained classification datasets including challenging CUB200-2011 bird, Flowers-102, and FGVC-Aircrafts datasets.

Zahra Anvari, Vassilis Athitsos

Haze can degrade the visibility and the image quality drastically, thus degrading the performance of computer vision tasks such as object detection. Single image dehazing is a challenging and ill-posed problem, despite being widely studied. Most existing methods assume that haze has a uniform/homogeneous distribution and haze can have a single color, i.e. grayish white color similar to smoke, while in reality haze can be distributed non-uniformly with different patterns and colors. In this paper, we focus on haze created by sunlight as it is one of the most prevalent type of haze in the wild. Sunlight can generate non-uniformly distributed haze with drastic density changes due to sun rays and also a spectrum of haze color due to sunlight color changes during the day. This presents a new challenge to image dehazing methods. For these methods to be practical, this problem needs to be addressed. To quantify the challenges and assess the performance of these methods, we present a sunlight haze benchmark dataset, Sun-Haze, containing 107 hazy images with different types of haze created by sunlight having a variety of intensity and color. We evaluate a representative set of state-of-the-art image dehazing methods on this benchmark dataset in terms of standard metrics such as PSNR, SSIM, CIEDE2000, PI and NIQE. This uncovers the limitation of the current methods, and questions their underlying assumptions as well as their practicality.

Zahra Anvari, Vassilis Athitsos

Single image de-hazing is a challenging problem, and it is far from solved. Most current solutions require paired image datasets that include both hazy images and their corresponding haze-free ground-truth images. However, in reality, lighting conditions and other factors can produce a range of haze-free images that can serve as ground truth for a hazy image, and a single ground truth image cannot capture that range. This limits the scalability and practicality of paired image datasets in real-world applications. In this paper, we focus on unpaired single image de-hazing and we do not rely on the ground truth image or physical scattering model. We reduce the image de-hazing problem to an image-to-image translation problem and propose a dehazing Global-Local Cycle-consistent Generative Adversarial Network (Dehaze-GLCGAN). Generator network of Dehaze-GLCGAN combines an encoder-decoder architecture with residual blocks to better recover the haze free scene. We also employ a global-local discriminator structure to deal with spatially varying haze. Through ablation study, we demonstrate the effectiveness of different factors in the performance of the proposed network. Our extensive experiments over three benchmark datasets show that our network outperforms previous work in terms of PSNR and SSIM while being trained on smaller amount of data compared to other methods.

Reza Ghoddoosian, Marnim Galib, Vassilis Athitsos

Drowsiness can put lives of many drivers and workers in danger. It is important to design practical and easy-to-deploy real-world systems to detect the onset of drowsiness.In this paper, we address early drowsiness detection, which can provide early alerts and offer subjects ample time to react. We present a large and public real-life dataset of 60 subjects, with video segments labeled as alert, low vigilant, or drowsy. This dataset consists of around 30 hours of video, with contents ranging from subtle signs of drowsiness to more obvious ones. We also benchmark a temporal model for our dataset, which has low computational and storage demands. The core of our proposed method is a Hierarchical Multiscale Long Short-Term Memory (HM-LSTM) network, that is fed by detected blink features in sequence. Our experiments demonstrate the relationship between the sequential blink features and drowsiness. In the experimental results, our baseline method produces higher accuracy than human judgment.

Srujana Gattupalli, Ashwin Ramesh Babu, James Robert Brady et al.

Hand keypoints detection and pose estimation has numerous applications in computer vision, but it is still an unsolved problem in many aspects. An application of hand keypoints detection is in performing cognitive assessments of a subject by observing the performance of that subject in physical tasks involving rapid finger motion. As a part of this work, we introduce a novel hand key-points benchmark dataset that consists of hand gestures recorded specifically for cognitive behavior monitoring. We explore the state of the art methods in hand keypoint detection and we provide quantitative evaluations for the performance of these methods on our dataset. In future, these results and our dataset can serve as a useful benchmark for hand keypoint recognition for rapid finger movements.

Wei Xiang, Dong-Qing Zhang, Heather Yu et al.

SSD is one of the state-of-the-art object detection algorithms, and it combines high detection accuracy with real-time speed. However, it is widely recognized that SSD is less accurate in detecting small objects compared to large objects, because it ignores the context from outside the proposal boxes. In this paper, we present CSSD--a shorthand for context-aware single-shot multibox object detector. CSSD is built on top of SSD, with additional layers modeling multi-scale contexts. We describe two variants of CSSD, which differ in their context layers, using dilated convolution layers (DiCSSD) and deconvolution layers (DeCSSD) respectively. The experimental results show that the multi-scale context modeling significantly improves the detection accuracy. In addition, we study the relationship between effective receptive fields (ERFs) and the theoretical receptive fields (TRFs), particularly on a VGGNet. The empirical results further strengthen our conclusion that SSD coupled with context layers achieves better detection results especially for small objects ($+3.2\% {\rm AP}_{@0.5}$ on MS-COCO compared to the newest SSD), while maintaining comparable runtime performance.

Amir Ghaderi, Srujana Gattupalli, Dylan Ebert et al.

HTKS is a game-like cognitive assessment method, designed for children between four and eight years of age. During the HTKS assessment, a child responds to a sequence of requests, such as "touch your head" or "touch your toes". The cognitive challenge stems from the fact that the children are instructed to interpret these requests not literally, but by touching a different body part than the one stated. In prior work, we have developed the CogniLearn system, that captures data from subjects performing the HTKS game, and analyzes the motion of the subjects. In this paper we propose some specific improvements that make the motion analysis module more accurate. As a result of these improvements, the accuracy in recognizing cases where subjects touch their toes has gone from 76.46% in our previous work to 97.19% in this paper.

Amir Ghaderi, Vassilis Athitsos

Supervised learning of convolutional neural networks (CNNs) can require very large amounts of labeled data. Labeling thousands or millions of training examples can be extremely time consuming and costly. One direction towards addressing this problem is to create features from unlabeled data. In this paper we propose a new method for training a CNN, with no need for labeled instances. This method for unsupervised feature learning is then successfully applied to a challenging object recognition task. The proposed algorithm is relatively simple, but attains accuracy comparable to that of more sophisticated methods. The proposed method is significantly easier to train, compared to existing CNN methods, making fewer requirements on manually labeled training data. It is also shown to be resistant to overfitting. We provide results on some well-known datasets, namely STL-10, CIFAR-10, and CIFAR-100. The results show that our method provides competitive performance compared with existing alternative methods. Selective Convolutional Neural Network (S-CNN) is a simple and fast algorithm, it introduces a new way to do unsupervised feature learning, and it provides discriminative features which generalize well.

Srujana Gattupalli, Amir Ghaderi, Vassilis Athitsos

Human body pose estimation and hand detection are two important tasks for systems that perform computer vision-based sign language recognition(SLR). However, both tasks are challenging, especially when the input is color videos, with no depth information. Many algorithms have been proposed in the literature for these tasks, and some of the most successful recent algorithms are based on deep learning. In this paper, we introduce a dataset for human pose estimation for SLR domain. We evaluate the performance of two deep learning based pose estimation methods, by performing user-independent experiments on our dataset. We also perform transfer learning, and we obtain results that demonstrate that transfer learning can improve pose estimation accuracy. The dataset and results from these methods can create a useful baseline for future works.

Hugo Jair Escalante, Isabelle Guyon, Vassilis Athitsos et al.

This paper introduces principal motion components (PMC), a new method for one-shot gesture recognition. In the considered scenario a single training-video is available for each gesture to be recognized, which limits the application of traditional techniques (e.g., HMMs). In PMC, a 2D map of motion energy is obtained per each pair of consecutive frames in a video. Motion maps associated to a video are processed to obtain a PCA model, which is used for recognition under a reconstruction-error approach. The main benefits of the proposed approach are its simplicity, easiness of implementation, competitive performance and efficiency. We report experimental results in one-shot gesture recognition using the ChaLearn Gesture Dataset; a benchmark comprising more than 50,000 gestures, recorded as both RGB and depth video with a Kinect camera. Results obtained with PMC are competitive with alternative methods proposed for the same data set.