Robotic catching has traditionally focused on single-handed systems, which are limited in their ability to handle larger or more complex objects. In contrast, bimanual catching offers significant potential for improved dexterity and object handling but introduces new challenges in coordination and control. In this paper, we propose a novel framework for learning dexterous bimanual catching skills using Heterogeneous-Agent Reinforcement Learning (HARL). Our approach introduces an adversarial reward scheme, where a throw agent increases the difficulty of throws-adjusting speed-while a catch agent learns to coordinate both hands to catch objects under these evolving conditions. We evaluate the framework in simulated environments using 15 different objects, demonstrating robustness and versatility in handling diverse objects. Our method achieved approximately a$2 x$increase in catching reward compared to single-agent baselines across$\mathbf{1 5}$diverse objects.

Imbuing embodied agents with non-verbal behavior offers significant benefits for agent-human interactions. Despite extensive research on the creation of non-verbal behaviors, the field lacks a standardized benchmarking practice. Researchers rarely compare their findings with previous studies, and when they do, the comparisons are often not methodologically aligned. The GENEA Workshop 2025 aims to bring together the non-verbal behavior generation community to discuss major challenges and solutions in the field and determine the most effective ways to advance it.

This work addresses the challenges associated with the use of glosses in both Sign Language Translation (SLT) and Sign Language Production (SLP). While glosses have long been used as a bridge between sign language and spoken language, they come with two major limitations that impede the advancement of sign language systems. First, annotating the glosses is a labor-intensive and time-consuming process, which limits the scalability of datasets. Second, the glosses oversimplify sign language by stripping away its spatio-temporal dynamics, reducing complex signs to basic labels and missing the subtle movements essential for precise interpretation. To address these limitations, we introduce Universal Gloss-level Representation (UniGloR), a framework designed to capture the spatio-temporal features inherent in sign language, providing a more dynamic and detailed alternative to the use of the glosses. The core idea of UniGloR is simple yet effective: We derive dense spatiotemporal representations from sign keypoint sequences using self-supervised learning and seamlessly integrate them into SLT and SLP tasks. Our experiments in a keypoint-based setting demonstrate that UniGloR either outperforms or matches the performance of previous SLT and SLP methods on two widely-used datasets: PHOENIX14T and How2Sign. Code is available at https://github.com/eddie-euijun-hwang/UniGloR.

Large language models (LLMs) have recently received considerable attention as alternative solutions for task planning. However, comparing the performance of language-oriented task planners becomes difficult, and there exists a dearth of detailed exploration regarding the effects of various factors such as pre-trained model selection and prompt construction. To address this, we propose a benchmark system for automatically quantifying performance of task planning for home-service embodied agents. Task planners are tested on two pairs of datasets and simulators: 1) ALFRED and AI2-THOR, 2) an extension of Watch-And-Help and VirtualHome. Using the proposed benchmark system, we perform extensive experiments with LLMs and prompts, and explore several enhancements of the baseline planner. We expect that the proposed benchmark tool would accelerate the development of language-oriented task planners.

Non-verbal behavior offers significant benefits for embodied agents in human interactions. Despite extensive research on the creation of non-verbal behaviors, the field lacks a standardized benchmarking practice. Researchers rarely compare their findings with previous studies, and when they do, the comparisons are often not aligned with other methodologies. The GENEA Workshop 2024 aims to unite the community to discuss major challenges and solutions, and to determine the most effective ways to advance the field.

This article reports on the second GENEA Challenge to benchmark data-driven automatic co-speech gesture generation. Participating teams used the same speech and motion dataset to build gesture-generation systems. Motion generated by all these systems was rendered to video using a standardised visualisation pipeline and evaluated in several large, crowdsourced user studies. Unlike when comparing different research articles, differences in results are here only due to differences between methods, enabling direct comparison between systems. The dataset was based on 18 hours of full-body motion capture, including fingers, of different persons engaging in a dyadic conversation. Ten teams participated in the challenge across two tiers: full-body and upper-body gesticulation. For each tier, we evaluated both the human-likeness of the gesture motion and its appropriateness for the specific speech signal. Our evaluations decouple human-likeness from gesture appropriateness, which has been a difficult problem in the field. The evaluation results show some synthetic gesture conditions being rated as significantly more human-like than 3D human motion capture. To the best of our knowledge, this has not been demonstrated before. On the other hand, all synthetic motion is found to be vastly less appropriate for the speech than the original motion-capture recordings. We also find that conventional objective metrics do not correlate well with subjective human-likeness ratings in this large evaluation. The one exception is the Fréchet gesture distance (FGD), which achieves a Kendall’s tau rank correlation of around -0.5. Based on the challenge results we formulate numerous recommendations for system building and evaluation.

We investigate the potential of the belief-desire-intention (BDI) model for enhancing proactive action planning and transparency in large language models (LLMs). Our proposed method, BDIPrompting, integrates the knowledge representation framework of the BDI model into prompt design. This allows agents to generate motivational and goal-directed service plans proactively while offering human users insights into the rationale behind the decision-making process. Through preliminary experiments with OpenAI’s GPT-4, we highlight the effectiveness of our approach in planning motivational actions and providing improved explanations during human-agent interactions.

Recent complicated problems require large-scale datasets and complex model architectures, however, it is difficult to train such large networks due to high computational issues. Significant efforts have been made to make the training more efficient such as momentum, learning rate scheduling, weight regularization, and meta-learning. Based on our observations on 1) high correlation between past eights and future weights, 2) conditions for beneficial weight prediction, and 3) feasibility of weight prediction, we propose a more general framework by intermittently skipping a handful of epochs by periodically forecasting near future weights, i.e., a Weight Nowcaster Network (WNN). As an add-on module, WNN predicts the future weights to make the learning process faster regardless of tasks and architectures. Experimental results show that WNN can significantly save actual time cost for training with an additional marginal time to train WNN. We validate the generalization capability of WNN under various tasks, and demonstrate that it works well even for unseen tasks. The code and pre-trained model are available at https://github.com/jjh6297/WNN.

Synthesizing realistic co-speech gestures is an important and yet unsolved problem for creating believable motions that can drive a humanoid robot to interact and communicate with human users. Such capability will improve the impressions of the robots by human users and will find applications in education, training, and medical services. One challenge in learning the co-speech gesture model is that there may be multiple viable gesture motions for the same speech utterance. The deterministic regression methods can not resolve the conflicting samples and may produce over-smoothed or damped motions. We proposed a two-stage model to address this uncertainty issue in gesture synthesis by modeling the gesture segments as discrete latent codes. Our method utilizes RQ- VAE in the first stage to learn a discrete codebook consisting of gesture tokens from training data. In the second stage, a two-level autoregressive transformer model is used to learn the prior distribution of residual codes conditioned on input speech context. Since the inference is formulated as token sampling, multiple gesture sequences could be generated given the same speech input using top-k sampling. The quantitative results and the user study showed the proposed method outperforms the previous methods and is able to generate realistic and diverse gesture motions.

Nowadays, Deep Learning-powered generative models are able to generate new synthetic samples nearly indistinguishable from natural data. The development of such systems necessarily involves the design of evaluation protocols to assess their performance. Quantitative objective metrics, such as Fréchet distance, in addition to human-centered subjective surveys, have become a standard for evaluating generative algorithms. Although motion generation is a popular research field, only a few works addressed the problem of the design and validation of a robust objective evaluation metric for motion-generative models. These previous works proposed to degrade ground truth motion samples with synthetic noises (e.g., Gaussian, Salt& Pepper) and studied the behavior of the proposed metric. However, this degradation does not mimic common motion artifacts produced by generative models. In this work, we propose (1) to validate Fréchet distance-based objective metrics on motion datasets degraded by two realistic motion artifacts, foot skating and over-smoothing, often found in motion synthesis results, and (2) a Fréchet Motion Distance (FMD), using Transformer-based feature extractor, able to capture the motion artifacts and also robust towards the variation of motion length.

Automatically generating character motion is one of the technologies required for virtual reality, graphics, and robotics. Motion synthesis with deep learning is an emerging research topic. A key component of the development of such an algorithm involves the design of a proper objective metric to evaluate the quality and diversity of the synthesized motion dataset, two key factors of the performance of generative models. The Fréchet distance is nowadays a common method to assess this performance. In the motion generation field, the validation of such evaluation methods relies on the computation of the Fréchet distance between embeddings of the ground truth dataset and motion samples polluted by synthetic noise to mimic the artifacts produced by generative algorithms. However, the synthetic noise degradation does not fully represent motion perturbations that are commonly perceived. One of these artifacts is foot skating: the unnatural foot slides on the ground during locomotion. In this work-in-progress paper, we tested how well the Fréchet Motion Distance (FMD), which was proposed in previous works, is able to measure foot skating artifacts, and we found that FMD is not able to measure efficiently the intensity of the skating degradation.

Embodied agents benefit from using non-verbal behavior when communicating with humans. Despite several decades of non-verbal behavior-generation research, there is currently no well-developed benchmarking culture in the field. For example, most researchers do not compare their outcomes with previous work, and if they do, they often do so in their own way which frequently is incompatible with others. With the GENEA Workshop 2022, we aim to bring the community together to discuss key challenges and solutions, and find the most appropriate ways to move the field forward.

This paper reports on the second GENEA Challenge to benchmark data-driven automatic co-speech gesture generation. Participating teams used the same speech and motion dataset to build gesture-generation systems. Motion generated by all these systems was rendered to video using a standardised visualisation pipeline and evaluated in several large, crowdsourced user studies. Unlike when comparing different research papers, differences in results are here only due to differences between methods, enabling direct comparison between systems. This year’s dataset was based on 18 hours of full-body motion capture, including fingers, of different persons engaging in dyadic conversation. Ten teams participated in the challenge across two tiers: full-body and upper-body gesticulation. For each tier we evaluated both the human-likeness of the gesture motion and its appropriateness for the specific speech signal. Our evaluations decouple human-likeness from gesture appropriateness, which previously was a major challenge in the field.

Modeling and generating realistic human gesture animations from speech audios has great impacts on creating a believable virtual human that can interact with human users and mimic real-world face-to-face communications. Large-scale datasets are essential in data-driven research, but creating multi-modal gesture datasets with 3D gesture motions and corresponding speech audios is either expensive to create via traditional workflow such as mocap, or producing subpar results via pose estimations from in-the-wild videos. As a result of such limitations, existing gesture datasets either suffer from shorter duration or lower animation quality, making them less ideal for training gesture synthesis models.

In many research areas, for example motion and gesture generation, objective measures alone do not provide an accurate impression of key stimulus traits such as perceived quality or appropriateness. The gold standard is instead to evaluate these aspects through user studies, especially subjective evaluations of video stimuli. Common evaluation paradigms either present individual stimuli to be scored on Likert-type scales, or ask users to compare and rate videos in a pairwise fashion. However, the time and resources required for such evaluations scale poorly as the number of conditions to be compared increases. Building on standards used for evaluating the quality of multimedia codecs, this paper instead introduces a framework for granular rating of multiple comparable videos in parallel. This methodology essentially analyses all condition pairs at once. Our contributions are 1) a proposed framework, called HEMVIP, for parallel and granular evaluation of multiple video stimuli and 2) a validation study confirming that results obtained using the tool are in close agreement with results of prior studies using conventional multiple pairwise comparisons.

Embodied agents benefit from using non-verbal behavior when communicating with humans. Despite several decades of non-verbal behavior-generation research, there is currently no well-developed benchmarking culture in the field. For example, most researchers do not compare their outcomes with previous work, and if they do, they often do so in their own way which frequently is incompatible with others. With the GENEA Workshop 2021, we aim to bring the community together to discuss key challenges and solutions, and find the most appropriate ways to move the field forward.

Co-speech gestures, gestures that accompany speech, play an important role in human communication. Automatic co-speech gesture generation is thus a key enabling technology for embodied conversational agents (ECAs), since humans expect ECAs to be capable of multi-modal communication. Research into gesture generation is rapidly gravitating towards data-driven methods. Unfortunately, individual research efforts in the field are difficult to compare: there are no established benchmarks, and each study tends to use its own dataset, motion visualisation, and evaluation methodology. To address this situation, we launched the GENEA Challenge, a gesture-generation challenge wherein participating teams built automatic gesture-generation systems on a common dataset, and the resulting systems were evaluated in parallel in a large, crowdsourced user study using the same motion-rendering pipeline. Since differences in evaluation outcomes between systems now are solely attributable to differences between the motion-generation methods, this enables benchmarking recent approaches against one another in order to get a better impression of the state of the art in the field. This paper reports on the purpose, design, results, and implications of our challenge.

Non-verbal behavior is essential for embodied agents like social robots, virtual avatars, and digital humans. Existing behavior authoring approaches including keyframe animation and motion capture are too expensive to use when there are numerous utterances requiring gestures. Automatic generation methods show promising results, but their output quality is not satisfactory yet, and it is hard to modify outputs as a gesture designer wants. We introduce a new gesture generation toolkit, named SGToolkit, which gives a higher quality output than automatic methods and is more efficient than manual authoring. For the toolkit, we propose a neural generative model that synthesizes gestures from speech and accommodates fine-level pose controls and coarse-level style controls from users. The user study with 24 participants showed that the toolkit is favorable over manual authoring, and the generated gestures were also human-like and appropriate to input speech. The SGToolkit is platform agnostic, and the code is available at https://github.com/ai4r/SGToolkit.

Near-surface multi-finger tracking (NMFT) technology expands the input space of touchscreens by enabling novel interactions such as mid-air and finger-aware interactions. We present DeepFisheye, a practical NMFT solution for mobile devices, that utilizes a fisheye camera attached at the bottom of a touchscreen. DeepFisheye acquires the image of an interacting hand positioned above the touchscreen using the camera and employs deep learning to estimate the 3D position of each fingertip. We created two new hand pose datasets comprising fisheye images, on which our network was trained. We evaluated DeepFisheye's performance for three device sizes. DeepFisheye showed average errors with approximate value of 20 mm for fingertip tracking across the different device sizes. Additionally, we created simple rule-based classifiers that estimate the contact finger and hand posture from DeepFisheye's output. The contact finger and hand posture classifiers showed accuracy of approximately 83 and 90%, respectively, across the device sizes.

For human-like agents, including virtual avatars and social robots, making proper gestures while speaking is crucial in human-agent interaction. Co-speech gestures enhance interaction experiences and make the agents look alive. However, it is difficult to generate human-like gestures due to the lack of understanding of how people gesture. Data-driven approaches attempt to learn gesticulation skills from human demonstrations, but the ambiguous and individual nature of gestures hinders learning. In this paper, we present an automatic gesture generation model that uses the multimodal context of speech text, audio, and speaker identity to reliably generate gestures. By incorporating a multimodal context and an adversarial training scheme, the proposed model outputs gestures that are human-like and that match with speech content and rhythm. We also introduce a new quantitative evaluation metric for gesture generation models. Experiments with the introduced metric and subjective human evaluation showed that the proposed gesture generation model is better than existing end-to-end generation models. We further confirm that our model is able to work with synthesized audio in a scenario where contexts are constrained, and show that different gesture styles can be generated for the same speech by specifying different speaker identities in the style embedding space that is learned from videos of various speakers. All the code and data is available at https://github.com/ai4r/Gesture-Generation-from-Trimodal-Context.

Co-speech gestures enhance interaction experiences between humans as well as between humans and robots. Most existing robots use rule-based speech-gesture association, but this requires human labor and prior knowledge of experts to be implemented. We present a learning-based co-speech gesture generation that is learned from 52 h of TED talks. The proposed end-to-end neural network model consists of an encoder for speech text understanding and a decoder to generate a sequence of gestures. The model successfully produces various gestures including iconic, metaphoric, deictic, and beat gestures. In a subjective evaluation, participants reported that the gestures were human-like and matched the speech content. We also demonstrate a co-speech gesture with a NAO robot working in real time.

This paper describes a novel RGB-D-based visual target tracking method for person-following robots. We enhance a single-object tracker, which combines RGB and depth information, by exploiting two different types of distracters. First set of distracters includes objects existing near-by the target, and the other set is for objects looking similar to the target. The proposed algorithm reduces tracking drifts and wrong target re-identification by exploiting the distracters. Experiments on real-world video sequences demonstrating a person-following problem show a significant improvement over the method without tracking distracters and state-of-the-art RGB-based trackers. A mobile robot following a person is tested in real environment.

This paper describes the reason that perception technology in HRI has not given high performance enough to be used for commercial service robots. As a practical solution for better performance of perception technology, we propose a perception framework with a dedicated perception engine called a perception demon. The demon in the proposed framework constantly collects evidences and analyses them better by combining various types of individual perception components. The proposed framework enables robot makers to easily get more reliable information on humans without concerns about optimizing perception components to their robots.

In everyday lives, so many events happen around us. Although not all of them are important to us, we have to process them all to be able to isolate the significant events. Consciousness is a constant awareness of the environment, which can be acquired only by such an exhaustive approach. As an autonomous agent, a robot is also required to be conscious, which is not yet properly realized. In this paper, we discuss the strategy to organize various perception technologies to achieve the consciousness of the robots. Especially, an independent process that constantly monitors its surroundings without the intervention of higher processes is proposed. This process, which is called a perception daemon, is discriminated from the traditional component-based approach, where individual perception technology is provided as a passive function so that the service applications should call whenever it is necessary.

This paper presents a person following robot equipped a RGB-D imaging sensor. We introduce three modules of a visual target tracking, target detection, and robot control. For the tracking module, distracters existing near-by the target are explicitly tracked to support target tracking. Preliminary tests showed that the robot robustly follows the target in uncontrolled environment with cluttered background and uneven illumination.

This paper presents a character segmentation method to address automatic number plate recognition problem. The method considered pixel intensity, character appearance, and arrangement of characters altogether to segment character regions. The method firstly discovers candidate blobs of characters by using connected component analysis and appearance-based character detection. A character recognizer is used for removing redundant and noisy blobs. Then, a trained classifier selects character blobs among the candidates by examining arrangement of the blobs. Experimental results show an achievement of 98.3% of segmentation rate, which prove the effectiveness of our method.

A character segmentation algorithm for automatic license plate recognition is presented in this paper. Character regions are selected through binarization, connected component analysis, and character recognition. A blob analysis operation excludes noisy blobs, merges fragmented blobs, and splits clumped blobs. A character segmentation module achieved an accuracy rate of 97.2%. The recognition accuracy of the complete system with license plate localization was 90.9%. In depth analysis of failure cases is also provided for better understanding of the algorithm and a future development direction.

This paper investigates image transformations as preprocessing steps that can be applied toward a state-of-the-art age classification framework of manifold learning. We report on the experimental results of four different preprocessing methods in terms of classification accuracy using a large training and test face database. The use of histograms of oriented gradient (HOG) descriptors increases the classification accuracy from 46 to 75%. Robustness to the rotation and translation is also tested.

A mobile phone with a camera enabled people to capture moments to remember in the right time and place. Due to a limited user interface, however, a mobile phone is not yet a platform for enjoying the captured moments. We explored possible application scenarios for promoting utilization of user-created photos on a mobile phone. For the realization of the scenarios, we designed context-aware photo selection algorithms that take into consideration mobile phone contexts such as the current location and recent calls. A user study was conducted with a mobile phone prototype for the evaluation of the photo selection algorithms and also for user feedback about the photo consumption scenarios.

We present a real-time simulation of spray painting incorporated into a VR environment as an alternative training system for ship-building industries. The system allows the user to try out a painting work on life-size structures with a spray gun. Our goal is to provide a trainee realistic painting experience in real-time as well as to represent the thickness of the deposited paint on the surface for evaluation of his performance. The Gaussian model is used for a painting deposition model, and texture mapping technique is utilized to provide efficient visual feedback. We also present effective collision detection methods for a volume of spray paint particles