Juan Ye

Ruilong Dan, Yunxiang Li, Yijie Wang et al.

Precise and rapid categorization of images in the B-scan ultrasound modality is vital for diagnosing ocular diseases. Nevertheless, distinguishing various diseases in ultrasound still challenges experienced ophthalmologists. Thus a novel contrastive disentangled network (CDNet) is developed in this work, aiming to tackle the fine-grained image categorization (FGIC) challenges of ocular abnormalities in ultrasound images, including intraocular tumor (IOT), retinal detachment (RD), posterior scleral staphyloma (PSS), and vitreous hemorrhage (VH). Three essential components of CDNet are the weakly-supervised lesion localization module (WSLL), contrastive multi-zoom (CMZ) strategy, and hyperspherical contrastive disentangled loss (HCD-Loss), respectively. These components facilitate feature disentanglement for fine-grained recognition in both the input and output aspects. The proposed CDNet is validated on our ZJU Ocular Ultrasound Dataset (ZJUOUSD), consisting of 5213 samples. Furthermore, the generalization ability of CDNet is validated on two public and widely-used chest X-ray FGIC benchmarks. Quantitative and qualitative results demonstrate the efficacy of our proposed CDNet, which achieves state-of-the-art performance in the FGIC task. Code is available at: https://github.com/ZeroOneGame/CDNet-for-OUS-FGIC .

Zhaojie Fang, Xiao Yu, Guanyu Zhou et al.

Ultra-Wide-Field Fluorescein Angiography (UWF-FA) enables precise identification of ocular diseases using sodium fluorescein, which can be potentially harmful. Existing research has developed methods to generate UWF-FA from Ultra-Wide-Field Scanning Laser Ophthalmoscopy (UWF-SLO) to reduce the adverse reactions associated with injections. However, these methods have been less effective in producing high-quality late-phase UWF-FA, particularly in lesion areas and fine details. Two primary challenges hinder the generation of high-quality late-phase UWF-FA: the scarcity of paired UWF-SLO and early/late-phase UWF-FA datasets, and the need for realistic generation at lesion sites and potential blood leakage regions. This study introduces an improved latent diffusion model framework to generate high-quality late-phase UWF-FA from limited paired UWF images. To address the challenges as mentioned earlier, our approach employs a module utilizing Cross-temporal Regional Difference Loss, which encourages the model to focus on the differences between early and late phases. Additionally, we introduce a low-frequency enhanced noise strategy in the diffusion forward process to improve the realism of medical images. To further enhance the mapping capability of the variational autoencoder module, especially with limited datasets, we implement a Gated Convolutional Encoder to extract additional information from conditional images. Our Latent Diffusion Model for Ultra-Wide-Field Late-Phase Fluorescein Angiography (LPUWF-LDM) effectively reconstructs fine details in late-phase UWF-FA and achieves state-of-the-art results compared to other existing methods when working with limited datasets. Our source code is available at: https://github.com/Tinysqua/****.

Yaxiong Lei, Shijing He, Mohamed Khamis et al.

In recent years we have witnessed an increasing number of interactive systems on handheld mobile devices which utilise gaze as a single or complementary interaction modality. This trend is driven by the enhanced computational power of these devices, higher resolution and capacity of their cameras, and improved gaze estimation accuracy obtained from advanced machine learning techniques, especially in deep learning. As the literature is fast progressing, there is a pressing need to review the state of the art, delineate the boundary, and identify the key research challenges and opportunities in gaze estimation and interaction. This paper aims to serve this purpose by presenting an end-to-end holistic view in this area, from gaze capturing sensors, to gaze estimation workflows, to deep learning techniques, and to gaze interactive applications.

Chengfeng Zhou, Songchang Chen, Chenming Xu et al.

Since data scarcity and data heterogeneity are prevailing for medical images, well-trained Convolutional Neural Networks (CNNs) using previous normalization methods may perform poorly when deployed to a new site. However, a reliable model for real-world clinical applications should be able to generalize well both on in-distribution (IND) and out-of-distribution (OOD) data (e.g., the new site data). In this study, we present a novel normalization technique called window normalization (WIN) to improve the model generalization on heterogeneous medical images, which is a simple yet effective alternative to existing normalization methods. Specifically, WIN perturbs the normalizing statistics with the local statistics computed on the window of features. This feature-level augmentation technique regularizes the models well and improves their OOD generalization significantly. Taking its advantage, we propose a novel self-distillation method called WIN-WIN for classification tasks. WIN-WIN is easily implemented with twice forward passes and a consistency constraint, which can be a simple extension for existing methods. Extensive experimental results on various tasks (6 tasks) and datasets (24 datasets) demonstrate the generality and effectiveness of our methods.

Yuheng Wang, Juan Ye, David L. Borchers

Passive acoustic monitoring can be an effective way of monitoring wildlife populations that are acoustically active but difficult to survey visually. Digital recorders allow surveyors to gather large volumes of data at low cost, but identifying target species vocalisations in these data is non-trivial. Machine learning (ML) methods are often used to do the identification. They can process large volumes of data quickly, but they do not detect all vocalisations and they do generate some false positives (vocalisations that are not from the target species). Existing wildlife abundance survey methods have been designed specifically to deal with the first of these mistakes, but current methods of dealing with false positives are not well-developed. They do not take account of features of individual vocalisations, some of which are more likely to be false positives than others. We propose three methods for acoustic spatial capture-recapture inference that integrate individual-level measures of confidence from ML vocalisation identification into the likelihood and hence integrate ML uncertainty into inference. The methods include a mixture model in which species identity is a latent variable. We test the methods by simulation and find that in a scenario based on acoustic data from Hainan gibbons, in which ignoring false positives results in 17% positive bias, our methods give negligible bias and coverage probabilities that are close to the nominal 95% level.

Ruiquan Ge, Xiangyang Hu, Rungen Huang et al.

Survival prediction plays a crucial role in assisting clinicians with the development of cancer treatment protocols. Recent evidence shows that multimodal data can help in the diagnosis of cancer disease and improve survival prediction. Currently, deep learning-based approaches have experienced increasing success in survival prediction by integrating pathological images and gene expression data. However, most existing approaches overlook the intra-modality latent information and the complex inter-modality correlations. Furthermore, existing modalities do not fully exploit the immense representational capabilities of neural networks for feature aggregation and disregard the importance of relationships between features. Therefore, it is highly recommended to address these issues in order to enhance the prediction performance by proposing a novel deep learning-based method. We propose a novel framework named Two-stream Transformer-based Multimodal Fusion Network for survival prediction (TTMFN), which integrates pathological images and gene expression data. In TTMFN, we present a two-stream multimodal co-attention transformer module to take full advantage of the complex relationships between different modalities and the potential connections within the modalities. Additionally, we develop a multi-head attention pooling approach to effectively aggregate the feature representations of the two modalities. The experiment results on four datasets from The Cancer Genome Atlas demonstrate that TTMFN can achieve the best performance or competitive results compared to the state-of-the-art methods in predicting the overall survival of patients.

Zhanyun Lu, Renshu Gu, Huimin Cheng et al.

Medical image datasets in the real world are often unlabeled and imbalanced, and Semi-Supervised Object Detection (SSOD) can utilize unlabeled data to improve an object detector. However, existing approaches predominantly assumed that the unlabeled data and test data do not contain out-of-distribution (OOD) classes. The few open-set semi-supervised object detection methods have two weaknesses: first, the class imbalance is not considered; second, the OOD instances are distinguished and simply discarded during pseudo-labeling. In this paper, we consider the open-set semi-supervised object detection problem which leverages unlabeled data that contain OOD classes to improve object detection for medical images. Our study incorporates two key innovations: Category Control Embed (CCE) and out-of-distribution Detection Fusion Classifier (OODFC). CCE is designed to tackle dataset imbalance by constructing a Foreground information Library, while OODFC tackles open-set challenges by integrating the ``unknown'' information into basic pseudo-labels. Our method outperforms the state-of-the-art SSOD performance, achieving a 4.25 mAP improvement on the public Parasite dataset.

Martin Schiemer, Clemens JS Schaefer, Jayden Parker Vap et al.

Continual learning is a desirable feature in many modern machine learning applications, which allows in-field adaptation and updating, ranging from accommodating distribution shift, to fine-tuning, and to learning new tasks. For applications with privacy and low latency requirements, the compute and memory demands imposed by continual learning can be cost-prohibitive for resource-constraint edge platforms. Reducing computational precision through fully quantized training (FQT) simultaneously reduces memory footprint and increases compute efficiency for both training and inference. However, aggressive quantization especially integer FQT typically degrades model accuracy to unacceptable levels. In this paper, we propose a technique that leverages inexpensive Hadamard transforms to enable low-precision training with only integer matrix multiplications. We further determine which tensors need stochastic rounding and propose tiled matrix multiplication to enable low-bit width accumulators. We demonstrate the effectiveness of our technique on several human activity recognition datasets and CIFAR100 in a class incremental learning setting. We achieve less than 0.5% and 3% accuracy degradation while we quantize all matrix multiplications inputs down to 4-bits with 8-bit accumulators.

Yaxiong Lei, Xinya Gong, Shijing He et al.

As eye-tracking becomes increasingly common in modern mobile devices, the potential for hands-free, gaze-based interaction grows, but current gesture sets are largely expert-designed and often misaligned with how users naturally move their eyes. To address this gap, we introduce a two-phase methodology for developing intuitive gaze gestures. First, four co-design workshops with 20 non-expert participants generated 102 initial concepts. Next, four gaze interaction experts reviewed and refined these into a set of 32 gestures. We found that non-experts, after a brief introduction, intuitively anchor gestures in familiar metaphors and develop a compositional grammar; i.e., activation (dwell) + action (gaze gesture or blink), to ensure intentionality and mitigate the classic Midas Touch problem. Experts prioritized gestures that are ergonomically sound, aligned with natural saccades, and reliably distinguishable. The resulting user-grounded, expert-validated gesture set, along with actionable design principles, provides a foundation for developing intuitive, hands-free interfaces for gaze-enabled devices.

Saurav Jha, Martin Schiemer, Franco Zambonelli et al.

Sensor-based human activity recognition (HAR), i.e., the ability to discover human daily activity patterns from wearable or embedded sensors, is a key enabler for many real-world applications in smart homes, personal healthcare, and urban planning. However, with an increasing number of applications being deployed, an important question arises: how can a HAR system autonomously learn new activities over a long period of time without being re-engineered from scratch? This problem is known as continual learning and has been particularly popular in the domain of computer vision, where several techniques to attack it have been developed. This paper aims to assess to what extent such continual learning techniques can be applied to the HAR domain. To this end, we propose a general framework to evaluate the performance of such techniques on various types of commonly used HAR datasets. We then present a comprehensive empirical analysis of their computational cost and effectiveness of tackling HAR-specific challenges (i.e., sensor noise and labels' scarcity). The presented results uncover useful insights on their applicability and suggest future research directions for HAR systems. Our code, models and data are available at https://github.com/srvCodes/continual-learning-benchmark.

Yaxiong Lei, Rishab Talwar, Shijing He et al.

Conventional mobile eye-tracking maps gaze to static screen coordinates, failing to capture user attention when content is dynamic. As users pinch, zoom, and rotate images, static coordinates lose their semantic meaning relative to the underlying visual content. To address this methodological gap, we present \textit{GazeSync}, a reusable mobile system that synchronizes on-device gaze estimation with real-time image transformation matrices (scale, rotation, and translation). By logging gaze coordinates alongside precise UI states, GazeSync enables the accurate reconstruction of \textit{image-relative} attention patterns, decoupling visual attention from device interaction. We validate our end-to-end toolchain through a formative study involving guided manipulation, reading, and visual search tasks. Our results demonstrate GazeSync's ability to recover ground-truth gaze locations on transforming content, explicitly showing how it outperforms static baselines, while also surfacing critical boundaries regarding calibration drift and reconstruction fragility under compound manipulations.

Yaxiong Lei, Thomas Davies, Xinya Gong et al.

Large-scale mobile gaze estimation relies on in-the-wild datasets, yet unsupervised collection makes it difficult to verify whether participants truly foveate logged targets. Prior mobile protocols often use low-entropy validation (e.g., binary probes) that can be satisfied by guessing and may still allow peripheral viewing, introducing label noise. We present \textbf{GazeCode}, a recall-based verification paradigm for higher-confidence in-the-wild mobile gaze data collection that strengthens \emph{label validity} through a multi-digit recall task (reducing random success to $10^{-N}$) paired with anti-peripheral stimulus design (small, low-contrast, brief digits). The system logs synchronized front-camera video, IMU streams, and target events using high-resolution timestamps. In a formative study (N=3), we probe key parameters (opacity, duration) and directly test peripheral exploitability using an eccentricity-controlled \textit{RING} condition. Results show that low-opacity digits substantially reduce peripheral readability while remaining usable for attentive foveation, supporting the inference that correct recall corresponds to higher-confidence gaze labels. We conclude with actionable design guidelines for robust in-the-wild gaze data collection.

Yaxiong Lei, Hyochan Cho, Fergus Buchanan et al.

Gaze gestures can provide hands free input on mobile devices, but practical use requires (i) gestures users can learn and recall and (ii) recognition models that are efficient enough for on-device deployment. We present an end-to-end pipeline using commodity ARKit head/eye transforms and a scaffolded guidance-to-recall protocol grounded in learning theory. In a pilot feasibility study (N=4 participants; 240 trials; controlled single-session setting), we benchmark a compact time-series model (TinyHAR) against deeper baselines (DeepConvLSTM, SA-HAR) on 5-way gesture recognition and 4-way user identification. TinyHAR achieves strong performance in this pilot benchmark (Macro F1 = 0.960 for gesture recognition; Macro F1 = 0.997 for user identification) while using only 46k parameters. A modality analysis further indicates that head pose dynamics are highly informative for mobile gaze gestures, highlighting embodied head--eye coordination as a key design consideration. Although the small sample size and controlled setting limit generalizability, these results indicate a potential direction for further investigation into on-device gaze gesture recognition.

Yaxiong Lei, Yuheng Wang, Fergus Buchanan et al.

Mobile gaze tracking involves inferring a user's gaze point or direction on a mobile device's screen from facial images captured by the device's front camera. While this technology inspires an increasing number of gaze-interaction applications, achieving consistent accuracy remains challenging due to dynamic user-device spatial relationships and varied motion conditions inherent in mobile contexts. This paper provides empirical evidence on how user mobility and behaviour affect mobile gaze tracking accuracy. We conduct two user studies collecting behaviour and gaze data under various motion conditions - from lying to maze navigation - and during different interaction tasks. Quantitative analysis has revealed behavioural regularities among daily tasks and identified head distance, head pose, and device orientation as key factors affecting accuracy, with errors increasing by up to 48.91% in dynamic conditions compared to static ones. These findings highlight the need for more robust, adaptive eye-tracking systems that account for head movements and device deflection to maintain accuracy across diverse mobile contexts.

Yaxiong Lei, Mingyue Zhao, Yuheng Wang et al.

Mobile gaze tracking faces a fundamental challenge: maintaining accuracy as users naturally change their postures and device orientations. Traditional calibration approaches, like one-off, fail to adapt to these dynamic conditions, leading to degraded performance over time. We present MAC-Gaze, a Motion-Aware continual Calibration approach that leverages smartphone Inertial measurement unit (IMU) sensors and continual learning techniques to automatically detect changes in user motion states and update the gaze tracking model accordingly. Our system integrates a pre-trained visual gaze estimator and an IMU-based activity recognition model with a clustering-based hybrid decision-making mechanism that triggers recalibration when motion patterns deviate significantly from previously encountered states. To enable accumulative learning of new motion conditions while mitigating catastrophic forgetting, we employ replay-based continual learning, allowing the model to maintain performance across previously encountered motion conditions. We evaluate our system through extensive experiments on the publicly available RGBDGaze dataset and our own 10-hour multimodal MotionGaze dataset (481K+ images, 800K+ IMU readings), encompassing a wide range of postures under various motion conditions including sitting, standing, lying, and walking. Results demonstrate that our method reduces gaze estimation error by 19.9% on RGBDGaze (from 1.73 cm to 1.41 cm) and by 31.7% on MotionGaze (from 2.81 cm to 1.92 cm) compared to traditional calibration approaches. Our framework provides a robust solution for maintaining gaze estimation accuracy in mobile scenarios.

Yiming Bao, Jun Wang, Tong Li et al.

The early diagnosis and screening of glaucoma are important for patients to receive treatment in time and maintain eyesight. Nowadays, deep learning (DL) based models have been successfully used for computer-aided diagnosis (CAD) of glaucoma from retina fundus images. However, a DL model pre-trained using a dataset from one hospital center may have poor performance on a dataset from another new hospital center and therefore its applications in the real scene are limited. In this paper, we propose a self-adaptive transfer learning (SATL) strategy to fill the domain gap between multicenter datasets. Specifically, the encoder of a DL model that is pre-trained on the source domain is used to initialize the encoder of a reconstruction model. Then, the reconstruction model is trained using only unlabeled image data from the target domain, which makes the encoder in the model adapt itself to extract useful high-level features both for target domain images encoding and glaucoma classification, simultaneously. Experimental results demonstrate that the proposed SATL strategy is effective in the domain adaptation task between one private and two public glaucoma diagnosis datasets, i.e. pri-RFG, REFUGE, and LAG. Moreover, the proposed strategy is completely independent of the source domain data, which meets the real scene application and the privacy protection policy.

Saurav Jha, Martin Schiemer, Juan Ye

Given the growing trend of continual learning techniques for deep neural networks focusing on the domain of computer vision, there is a need to identify which of these generalizes well to other tasks such as human activity recognition (HAR). As recent methods have mostly been composed of loss regularization terms and memory replay, we provide a constituent-wise analysis of some prominent task-incremental learning techniques employing these on HAR datasets. We find that most regularization approaches lack substantial effect and provide an intuition of when they fail. Thus, we make the case that the development of continual learning algorithms should be motivated by rather diverse task domains.