The prevalence of user-generated content (UGC) on platforms such as YouTube and TikTok has rendered no-reference (NR) perceptual video quality assessment (VQA) vital for optimizing video delivery. Nonetheless, the characteristics of non-professional acquisition and the subsequent transcoding of UGC video on sharing platforms present significant challenges for NR-VQA. Although NR-VQA models attempt to infer mean opinion scores (MOS), their modeling of subjective scores for compressed content remains limited due to the absence of fine-grained perceptual annotations of artifact types. To address these challenges, we propose CAMP-VQA, a novel NR-VQA framework that exploits the semantic understanding capabilities of large vision-language models. Our approach introduces a quality-aware prompting mechanism that integrates video metadata (e.g., resolution, frame rate, bitrate) with key fragments extracted from inter-frame variations to guide the BLIP-2 pretraining approach in generating fine-grained quality captions. A unified architecture has been designed to model perceptual quality across three dimensions: semantic alignment, temporal characteristics, and spatial characteristics. These multimodal features are extracted and fused, then regressed to video quality scores. Extensive experiments on a wide variety of UGC datasets demonstrate that our model consistently outperforms existing NR-VQA methods, achieving improved accuracy without the need for costly manual fine-grained annotations. Our method achieves the best performance in terms of average rank and linear correlation (SRCC: 0.928, PLCC: 0.938) compared to state-of-the-art methods. The source code and trained models, along with a user-friendly demo, are available at: https://github.com/xinyiW915/CAMP-VQA.

Humans excel at spatial-temporal reasoning, effortlessly interpreting dynamic visual events from an egocentric viewpoint. However, whether multimodal large language models (MLLMs) can similarly understand the 4D world remains uncertain. This paper explores multimodal spatial-temporal reasoning from an egocentric perspective, aiming to equip MLLMs with human-like reasoning capabilities. To support this objective, we introduce Ego-ST Bench, a novel benchmark containing over 5,000 question-answer pairs across four categories, systematically evaluating spatial, temporal, and integrated spatial-temporal reasoning. Additionally, we propose ST-R1 training paradigm, a video-based reasoning model that incorporates reverse thinking into its reinforcement learning process, significantly enhancing performance. We combine long-chain-of-thought (long-CoT) supervised fine-tuning with Group Relative Policy Optimization (GRPO) reinforcement learning, achieving notable improvements with limited high-quality data. Ego-ST Bench and ST-R1 provide valuable insights and resources for advancing video-based spatial-temporal reasoning research.

With the rapid advancement of AI systems, their abilities to store, retrieve, and utilize information over the long term - referred to as long-term memory - have become increasingly significant. These capabilities are crucial for enhancing the performance of AI systems across a wide range of tasks. However, there is currently no comprehensive survey that systematically investigates AI's long-term memory capabilities, formulates a theoretical framework, and inspires the development of next-generation AI long-term memory systems. This paper begins by introducing the mechanisms of human long-term memory, then explores AI long-term memory mechanisms, establishing a mapping between the two. Based on the mapping relationships identified, we extend the current cognitive architectures and propose the Cognitive Architecture of Self-Adaptive Long-term Memory (SALM). SALM provides a theoretical framework for the practice of AI long-term memory and holds potential for guiding the creation of next-generation long-term memory driven AI systems. Finally, we delve into the future directions and application prospects of AI long-term memory.

Current state-of-the-art 3D reconstruction models face limitations in building extra-large scale outdoor scenes, primarily due to the lack of sufficiently large-scale and detailed datasets. In this paper, we present a extra-large fine-grained dataset with 10 billion points composed of 41,006 drone-captured high-resolution aerial images, covering 20 diverse and culturally significant scenes from worldwide locations such as Cambridge Uni main buildings, the Pyramids, and the Forbidden City Palace. Compared to existing datasets, ours offers significantly larger scale and higher detail, uniquely suited for fine-grained 3D applications. Each scene contains an accurate spatial layout and comprehensive structural information, supporting detailed 3D reconstruction tasks. By reconstructing environments using these detailed images, our dataset supports multiple applications, including outputs in the widely adopted COLMAP format, establishing a novel benchmark for evaluating state-of-the-art large-scale Gaussian Splatting methods.The dataset's flexibility encourages innovations and supports model plug-ins, paving the way for future 3D breakthroughs. All datasets and code will be open-sourced for community use.

Training a real-time gesture recognition model heavily relies on annotated data. However, manual data annotation is costly and demands substantial human effort. In order to address this challenge, we propose a framework that can automatically annotate gesture classes and identify their temporal ranges. Our framework consists of two key components: (1) a novel annotation model that leverages the Connectionist Temporal Classification (CTC) loss, and (2) a semi-supervised learning pipeline that enables the model to improve its performance by training on its own predictions, known as pseudo labels. These high-quality pseudo labels can also be used to enhance the accuracy of other downstream gesture recognition models. To evaluate our framework, we conducted experiments using two publicly available gesture datasets. Our ablation study demonstrates that our annotation model design surpasses the baseline in terms of both gesture classification accuracy (3–4 % improvement) and localization accuracy (71-75% improvement). Additionally, we illustrate that the pseudo-labeled dataset produced from the proposed framework significantly boosts the accuracy of a pre-trained downstream gesture recognition model by 11-18%. We believe that this annotation framework has immense potential to improve the training of downstream gesture recognition models using unlabeled datasets.

We depend on our own memory to encode, store, and retrieve our experiences. However, memory lapses can occur. One promising avenue for achieving memory augmentation is through the use of augmented reality head-mounted displays to capture and preserve egocentric videos, a practice commonly referred to as lifelogging. However, a significant challenge arises from the sheer volume of video data generated through lifelogging, as the current technology lacks the capability to encode and store such large amounts of data efficiently. Further, retrieving specific information from extensive video archives requires substantial computational power, further complicating the task of quickly accessing desired content. To address these challenges, we propose a memory augmentation agent that involves leveraging natural language encoding for video data and storing them in a vector database. This approach harnesses the power of large vision language models to perform the language encoding process. Additionally, we propose using large language models to facilitate natural language querying. Our agent underwent extensive evaluation using the QA-Ego4D dataset and achieved state-of-the-art results with a BLEU score of 8.3, outperforming conventional machine learning models that scored between 3.4 and 5.8. Additionally, we conducted a user study in which participants interacted with the human memory augmentation agent through episodic memory and open-ended questions. The results of this study show that the agent results in significantly better recall performance on episodic memory tasks compared to human participants. The results also highlight the agent’s practical applicability and user acceptance.

Providing users with accurate gestural interfaces, such as gesture recognition based on wrist-worn devices, is a key challenge in mixed reality. However, static machine learning processes in gesture recognition assume that training and test data come from the same underlying distribution. Unfortunately, in real-world applications involving gesture recognition, such as gesture recognition based on wrist-worn devices, the data distribution may change over time. We formulate this problem of adapting recognition models to new tasks, where new data patterns emerge, as open-world gesture recognition (OWGR). We propose the use of continual learning to enable machine learning models to be adaptive to new tasks without degrading performance on previously learned tasks. However, the process of exploring parameters for questions around when, and how, to train and deploy recognition models requires resource-intensive user studies may be impractical. To address this challenge, we propose a design engineering approach that enables offline analysis on a collected large-scale dataset by systematically examining various parameters and comparing different continual learning methods. Finally, we provide design guidelines to enhance the development of an open-world wrist-worn gesture recognition process.

Text entry is a critical capability for any modern computing experience, with lightweight augmented reality (AR) glasses being no exception. Designed for all-day wearability, a limitation of lightweight AR glass is the restriction to the inclusion of multiple cameras for extensive field of view in hand tracking. This constraint underscores the need for an additional input device. We propose a system to address this gap: a ring-based mid-air gesture typing technique, RingGesture, utilizing electrodes to mark the start and end of gesture trajectories and inertial measurement units (IMU) sensors for hand tracking. This method offers an intuitive experience similar to raycast-based mid-air gesture typing found in VR headsets, allowing for a seamless translation of hand movements into cursor navigation. To enhance both accuracy and input speed, we propose a novel deep-learning word prediction framework, Score Fusion, comprised of three key components: a) a word-gesture decoding model, b) a spatial spelling correction model, and c) a lightweight contextual language model. In contrast, this framework fuses the scores from the three models to predict the most likely words with higher precision. We conduct comparative and longitudinal studies to demonstrate two key findings: firstly, the overall effectiveness of RingGesture, which achieves an average text entry speed of 27.3 words per minute (WPM) and a peak performance of 47.9 WPM. Secondly, we highlight the superior performance of the Score Fusion framework, which offers a 28.2% improvement in uncorrected Character Error Rate over a conventional word prediction framework, Naive Correction, leading to a 55.2% improvement in text entry speed for RingGesture. Additionally, RingGesture received a System Usability Score of 83 signifying its excellent usability.

Text entry with word-gesture keyboards (WGK) is emerging as a popular method and becoming a key interaction for Extended Reality (XR). However, the diversity of interaction modes, keyboard sizes, and visual feedback in these environments introduces divergent word-gesture trajectory data patterns, thus leading to complexity in decoding trajectories into text. Template-matching decoding methods, such as SHARK2 [32], are commonly used for these WGK systems because they are easy to implement and configure. However, these methods are susceptible to decoding inaccuracies for noisy trajectories. While conventional neural-network-based decoders (neural decoders) trained on word-gesture trajectory data have been proposed to improve accuracy, they have their own limitations: they require extensive data for training and deep-learning expertise for implementation. To address these challenges, we propose a novel solution that combines ease of implementation with high decoding accuracy: a generalizable neural decoder enabled by pre-training on large-scale coarsely discretized word-gesture trajectories. This approach produces a ready-to-use WGK decoder that is generalizable across mid-air and on-surface WGK systems in augmented reality (AR) and virtual reality (VR), which is evident by a robust average Top-4 accuracy of 90.4% on four diverse datasets. It significantly outperforms SHARK2 with a 37.2% enhancement and surpasses the conventional neural decoder by 7.4%. Moreover, the Pre-trained Neural Decoder's size is only 4 MB after quantization, without sacrificing accuracy, and it can operate in real-time, executing in just 97 milliseconds on Quest 3.

Alphanumeric and special characters are essential during text entry. Text entry in virtual reality (VR) is usually performed on a virtual Qwerty keyboard to minimize the need to learn new layouts. As such, entering capitals, symbols, and numbers in VR is often a direct migration from a physical/touchscreen Qwerty keyboard-that is, using the mode-switching keys to switch between different types of characters and symbols. However, there are inherent differences between a keyboard in VR and a physical/touchscreen keyboard, and as such, a direct adaptation of mode-switching via switch keys may not be suitable for VR. The high flexibility afforded by VR opens up more possibilities for entering alphanumeric and special characters using the Qwerty layout. In this work, we designed two controller-based raycasting text entry methods for alphanumeric and special characters input (Layer-ButtonSwitch and Key-ButtonSwitch) and compared them with two other methods (Standard Qwerty Keyboard and Layer-PointSwitch) that were derived from physical and soft Qwerty keyboards. We explored the performance and user preference of these four methods via two user studies (one short-term and one prolonged use), where participants were instructed to input text containing alphanumeric and special characters. Our results show that Layer-ButtonSwitch led to the highest statistically significant performance, followed by Key-ButtonSwitch and Standard Qwerty Keyboard, while Layer-PointSwitch had the slowest speed. With continuous practice, participants' performance using Key-ButtonSwitch reached that of Layer-ButtonSwitch. Further, the results show that the key-level layout used in Key-ButtonSwitch led users to parallel mode switching and character input operations because this layout showed all characters on one layer. We distill three recommendations from the results that can help guide the design of text entry techniques for alphanumeric and special characters in VR.

Explainable AI (XAI) has established itself as an important component of AI-driven interactive systems. With Augmented Reality (AR) becoming more integrated in daily lives, the role of XAI also becomes essential in AR because end-users will frequently interact with intelligent services. However, it is unclear how to design effective XAI experiences for AR. We propose XAIR, a design framework that addresses when, what, and how to provide explanations of AI output in AR. The framework was based on a multi-disciplinary literature review of XAI and HCI research, a large-scale survey probing 500+ end-users’ preferences for AR-based explanations, and three workshops with 12 experts collecting their insights about XAI design in AR. XAIR’s utility and effectiveness was verified via a study with 10 designers and another study with 12 end-users. XAIR can provide guidelines for designers, inspiring them to identify new design opportunities and achieve effective XAI designs in AR.

We present a fast mid-air gesture keyboard for head-mounted optical see-through augmented reality (OST AR) that supports users in articulating word patterns by merely moving their own physical index finger in relation to a virtual keyboard plane without a need to indirectly control a visual 2D cursor on a keyboard plane. To realize this, we introduce a novel decoding method that directly translates users' three-dimensional fingertip gestural trajectories into their intended text. We evaluate the efficacy of the system in three studies that investigate various design aspects, such as immediate efficacy, accelerated learning, and whether it is possible to maintain performance without providing visual feedback. We find that the new 3D trajectory decoding design results in significant improvements in entry rates while maintaining low error rates. In addition, we demonstrate that users can maintain their performance even without fingertip and gesture trace visualization.

We present AdaptiKeyboard, a mid-air gesture keyboard that uses multi-objective Bayesian optimization to adaptively change layout size to simultaneously optimize speed and accuracy. Gesture keyboards are well suited for enabling mid-air text entry in augmented reality (AR) due to their relative robustness to articulation inaccuracy. However, transplanting gesture keyboards to AR involves a larger design and operational space compared to touchscreen interactions. One potential advantage of this larger design and operational space is that mid-air keyboards presented in AR can be more versatile than their touchscreen equivalents. A key component of a mid-air gesture keyboard is the layout size, which can be made adaptive in order to optimize text entry speed and accuracy at the individual user level. This adaptive personalization can refine the keyboard design to reflect the differences users exhibit in motor behaviors and personal preferences. In this paper, we propose a multi-objective Bayesian optimization approach for adapting the layout size of a mid-air gesture keyboard to individual users. We show that this process can deliver a 14.4% improvement in speed and a 13.8% improvement in accuracy relative to a baseline design with a constant size derived from the default system keyboard on the HoloLens 2.

We present KWickChat (Keyword Quick Chat): a multi-turn augmentative and alternative communication (AAC) dialogue system for nonspeaking individuals with motor disabilities. The central objective of KWickChat is to reduce the communication gap between nonspeaking and speaking partners by exploring a sentence-based text entry system that automatically generates suitable sentences for the nonspeaking partner based on keyword entry. The system is underpinned by a GPT-2 language model and leverages context information, including dialogue history and persona tags, to improve the quality of the generated responses. We evaluate the system by analyzing the functional design and decomposing it into key functions and parameters that are systematically investigated using envelope analysis. We pursue this methodology as a necessary precursor to evaluation with AAC users. Our results show that with word prediction and with a threshold word error rate of 0.65, the keystroke savings of the KWickChat system is around 71%. To complement the envelope analysis, we also recruited two human judges to evaluate the semantic consistency between 400 sentences generated by KWickChat and reference sentences. Both judges reported a median rating of 4 on a scale from 1 (very bad) to 5 (very good) for the best generated sentence in each exchange and achieved an inter-rater reliability of 0.92 across all 400 sentences judged.

In this paper we examine the task of key gesture spotting: accurate and timely online recognition of hand gestures. We specifically seek to address two key challenges faced by developers when integrating key gesture spotting functionality into their applications. These are: i) achieving high accuracy and zero or negative activation lag with single-time activation; and ii) avoiding the requirement for deep domain expertise in machine learning. We address the first challenge by proposing a key gesture spotting architecture consisting of a novel gesture classifier model and a novel single-time activation algorithm. This key gesture spotting architecture was evaluated on four separate hand skeleton gesture datasets, and achieved high recognition accuracy with early detection. We address the second challenge by encapsulating different data processing and augmentation strategies, as well as the proposed key gesture spotting architecture, into a graphical user interface and an application programming interface. Two user studies demonstrate that developers are able to efficiently construct custom recognizers using both the graphical user interface and the application programming interface.

Deep learning approaches deliver state-of-the-art performance in recognition of spatiotemporal human motion data. However, one of the main challenges in these recognition tasks is limited available training data. Insufficient training data results in over-fitting and data augmentation is one approach to address this challenge. Existing data augmentation strategies based on scaling, shifting and interpolating offer limited generalizability and typically require detailed inspection of the dataset as well as hundreds of GPU hours for hyperparameter optimization. In this paper, we present a novel automatic data augmentation model, the Imaginative Generative Adversarial Network (GAN), that approximates the distribution of the input data and samples new data from this distribution. It is automatic in that it requires no data inspection and little hyperparameter tuning and therefore it is a low-cost and low-effort approach to generate synthetic data. We demonstrate our approach on small-scale skeleton-based datasets with a comprehensive experimental analysis. Our results show that the augmentation strategy is fast to train and can improve classification accuracy for both conventional neural networks and state-of-the-art methods.

The eventual success of many AR and VR intelligent interactive systems relies on the ability to collect user motion data at large scale. Realistic simulation of human motion trajectories is a potential solution to this problem. Simulated user motion data can facilitate prototyping and speed up the design process. There are also potential benefits in augmenting training data for deep learning-based AR/VR applications to improve performance. However, the generation of realistic motion data is nontrivial. In this paper, we examine the specific challenge of simulating index finger movement data to inform mid-air gesture keyboard design. The mid-air gesture keyboard is deployed on an optical see-through display that allows the user to enter text by articulating word gesture patterns with their physical index finger in the vicinity of a visualized keyboard layout. We propose and compare four different approaches to simulating this type of motion data, including a Jerk-Minimization model, a Recurrent Neural Network (RNN)-based generative model, and a Generative Adversarial Network (GAN)-based model with two modes: style transfer and data alteration. We also introduce a procedure for validating the quality of the generated trajectories in terms of realism and diversity. The GAN-based model shows significant potential for generating synthetic motion trajectories to facilitate design and deep learning for advanced gesture keyboards deployed in AR and VR.