arXiv subjects

Jaehong Kim

Publications and source records attributed to Jaehong Kim.

34 records · Page 2Linked to original sources

arXiv2021

SGToolkit: An Interactive Gesture Authoring Toolkit for Embodied Conversational Agents

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.

cs.HC↗

arXiv2021

Cut-and-Paste Dataset Generation for Balancing Domain Gaps in Object Instance Detection

Training an object instance detector where only a few training object images are available is a challenging task. One solution is a cut-and-paste method that generates a training dataset by cutting object areas out of training images and pasting them onto other background images. A detector trained on a dataset generated with a cut-and-paste method suffers from the conventional domain shift problem, which stems from a discrepancy between the source domain (generated training dataset) and the target domain (real test dataset). Though state-of-the-art domain adaptation methods are able to reduce this gap, it is limited because they do not consider the difference of domain gaps of foreground and background. In this study, we present that the conventional domain gap can be divided into two sub-domain gaps for foreground and background. Then, we show that the original cut-and-paste approach suffers from a new domain gap problem, an unbalanced domain gaps, because it has two separate source domains for foreground and background, unlike the conventional domain shift problem. Then, we introduce an advanced cut-and-paste method to balance the unbalanced domain gaps by diversifying the foreground with GAN (generative adversarial network)-generated seed images and simplifying the background using image processing techniques. Experimental results show that our method is effective for balancing domain gaps and improving the accuracy of object instance detection in a cluttered indoor environment using only a few seed images. Furthermore, we show that balancing domain gaps can improve the detection accuracy of state-of-the-art domain adaptation methods.

cs.RO↗

arXiv2021

VOTE400(Voide Of The Elderly 400 Hours): A Speech Dataset to Study Voice Interface for Elderly-Care

This paper introduces a large-scale Korean speech dataset, called VOTE400, that can be used for analyzing and recognizing voices of the elderly people. The dataset includes about 300 hours of continuous dialog speech and 100 hours of read speech, both recorded by the elderly people aged 65 years or over. A preliminary experiment showed that speech recognition system trained with VOTE400 can outperform conventional systems in speech recognition of elderly people's voice. This work is a multi-organizational effort led by ETRI and MINDs Lab Inc. for the purpose of advancing the speech recognition performance of the elderly-care robots.

eess.AS↗

arXiv2020

AIR-Act2Act: Human-human interaction dataset for teaching non-verbal social behaviors to robots

To better interact with users, a social robot should understand the users' behavior, infer the intention, and respond appropriately. Machine learning is one way of implementing robot intelligence. It provides the ability to automatically learn and improve from experience instead of explicitly telling the robot what to do. Social skills can also be learned through watching human-human interaction videos. However, human-human interaction datasets are relatively scarce to learn interactions that occur in various situations. Moreover, we aim to use service robots in the elderly-care domain; however, there has been no interaction dataset collected for this domain. For this reason, we introduce a human-human interaction dataset for teaching non-verbal social behaviors to robots. It is the only interaction dataset that elderly people have participated in as performers. We recruited 100 elderly people and two college students to perform 10 interactions in an indoor environment. The entire dataset has 5,000 interaction samples, each of which contains depth maps, body indexes and 3D skeletal data that are captured with three Microsoft Kinect v2 cameras. In addition, we provide the joint angles of a humanoid NAO robot which are converted from the human behavior that robots need to learn. The dataset and useful python scripts are available for download at https://github.com/ai4r/AIR-Act2Act. It can be used to not only teach social skills to robots but also benchmark action recognition algorithms.

cs.RO↗

arXiv2020

Speech Gesture Generation from the Trimodal Context of Text, Audio, and Speaker Identity

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.

cs.GR↗

arXiv2020

Vector Control Algorithm Based on Different Current Control Switching Techniques for Ac Motor Drives

A comparative analysis of vector control scheme based on different current control switching pulses (HC, SPWM, DPWM and SVPWM) for the speed response of motor drive is analysed in this paper. The control system using different switching techniques, are comparatively simulated and analysed. Ac motor drives are progressively used in high-performance application industries due to small size, efficient performance, robust to torque response and high power to size ratio. A mathematical model of ac motor drives is presented in order to explain the numerical theory of motor drives. The vector control technique is utilized for efficient speed control of ac motor drive based on independent torque and air gap flux control. The study compares the total harmonic distortion contents of phase currents of ac motor drive and speed response in each case. The simulation result shows that total harmonic distortion across the phase current in SVPWM is less as compared to other switching techniques while the rise time in speed response across SVPWM technique is faster as compared to other switching methods. The simulation result of ac motor drives speed control is demonstrated in Matlab/Simulink 2018b.

eess.SY↗

arXiv2020

ETRI-Activity3D: A Large-Scale RGB-D Dataset for Robots to Recognize Daily Activities of the Elderly

Deep learning, based on which many modern algorithms operate, is well known to be data-hungry. In particular, the datasets appropriate for the intended application are difficult to obtain. To cope with this situation, we introduce a new dataset called ETRI-Activity3D, focusing on the daily activities of the elderly in robot-view. The major characteristics of the new dataset are as follows: 1) practical action categories that are selected from the close observation of the daily lives of the elderly; 2) realistic data collection, which reflects the robot's working environment and service situations; and 3) a large-scale dataset that overcomes the limitations of the current 3D activity analysis benchmark datasets. The proposed dataset contains 112,620 samples including RGB videos, depth maps, and skeleton sequences. During the data acquisition, 100 subjects were asked to perform 55 daily activities. Additionally, we propose a novel network called four-stream adaptive CNN (FSA-CNN). The proposed FSA-CNN has three main properties: robustness to spatio-temporal variations, input-adaptive activation function, and extension of the conventional two-stream approach. In the experiment section, we confirmed the superiority of the proposed FSA-CNN using NTU RGB+D and ETRI-Activity3D. Further, the domain difference between both groups of age was verified experimentally. Finally, the extension of FSA-CNN to deal with the multimodal data was investigated.

cs.RO↗

arXiv2019

Deep Asymmetric Networks with a Set of Node-wise Variant Activation Functions

This work presents deep asymmetric networks with a set of node-wise variant activation functions. The nodes' sensitivities are affected by activation function selections such that the nodes with smaller indices become increasingly more sensitive. As a result, features learned by the nodes are sorted by the node indices in the order of their importance. Asymmetric networks not only learn input features but also the importance of those features. Nodes of lesser importance in asymmetric networks can be pruned to reduce the complexity of the networks, and the pruned networks can be retrained without incurring performance losses. We validate the feature-sorting property using both shallow and deep asymmetric networks as well as deep asymmetric networks transferred from famous networks.

cs.LG↗

arXiv2018

Auto-Meta: Automated Gradient Based Meta Learner Search

Fully automating machine learning pipelines is one of the key challenges of current artificial intelligence research, since practical machine learning often requires costly and time-consuming human-powered processes such as model design, algorithm development, and hyperparameter tuning. In this paper, we verify that automated architecture search synergizes with the effect of gradient-based meta learning. We adopt the progressive neural architecture search \cite{liu:pnas_google:DBLP:journals/corr/abs-1712-00559} to find optimal architectures for meta-learners. The gradient based meta-learner whose architecture was automatically found achieved state-of-the-art results on the 5-shot 5-way Mini-ImageNet classification problem with $74.65\%$ accuracy, which is $11.54\%$ improvement over the result obtained by the first gradient-based meta-learner called MAML \cite{finn:maml:DBLP:conf/icml/FinnAL17}. To our best knowledge, this work is the first successful neural architecture search implementation in the context of meta learning.

cs.LG↗

arXiv2018

Neural Networks with Activation Networks

This work presents an adaptive activation method for neural networks that exploits the interdependency of features. Each pixel, node, and layer is assigned with a polynomial activation function, whose coefficients are provided by an auxiliary activation network. The activation of a feature depends on the features of neighboring pixels in a convolutional layer and other nodes in a dense layer. The dependency is learned from data by the activation networks. In our experiments, networks with activation networks provide significant performance improvement compared to the baseline networks on which they are built. The proposed method can be used to improve the network performance as an alternative to increasing the number of nodes and layers.

cs.CV↗

arXiv2018

Robots Learn Social Skills: End-to-End Learning of Co-Speech Gesture Generation for Humanoid Robots

Co-speech gestures enhance interaction experiences between humans as well as between humans and robots. 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.

cs.RO↗

arXiv2018

Doubly Nested Network for Resource-Efficient Inference

We propose doubly nested network(DNNet) where all neurons represent their own sub-models that solve the same task. Every sub-model is nested both layer-wise and channel-wise. While nesting sub-models layer-wise is straight-forward with deep-supervision as proposed in \cite{xie2015holistically}, channel-wise nesting has not been explored in the literature to our best knowledge. Channel-wise nesting is non-trivial as neurons between consecutive layers are all connected to each other. In this work, we introduce a technique to solve this problem by sorting channels topologically and connecting neurons accordingly. For the purpose, channel-causal convolutions are used. Slicing doubly nested network gives a working sub-network. The most notable application of our proposed network structure with slicing operation is resource-efficient inference. At test time, computing resources such as time and memory available for running the prediction algorithm can significantly vary across devices and applications. Given a budget constraint, we can slice the network accordingly and use a sub-model for inference within budget, requiring no additional computation such as training or fine-tuning after deployment. We demonstrate the effectiveness of our approach in several practical scenarios of utilizing available resource efficiently.

cs.LG↗

arXiv2017

Continual Learning with Deep Generative Replay

Attempts to train a comprehensive artificial intelligence capable of solving multiple tasks have been impeded by a chronic problem called catastrophic forgetting. Although simply replaying all previous data alleviates the problem, it requires large memory and even worse, often infeasible in real world applications where the access to past data is limited. Inspired by the generative nature of hippocampus as a short-term memory system in primate brain, we propose the Deep Generative Replay, a novel framework with a cooperative dual model architecture consisting of a deep generative model ("generator") and a task solving model ("solver"). With only these two models, training data for previous tasks can easily be sampled and interleaved with those for a new task. We test our methods in several sequential learning settings involving image classification tasks.

cs.AI↗

arXiv2013

Modulation Coding for Flash Memories

The aggressive scaling down of flash memories has threatened data reliability since the scaling down of cell sizes gives rise to more serious degradation mechanisms such as cell-to-cell interference and lateral charge spreading. The effect of these mechanisms has pattern dependency and some data patterns are more vulnerable than other ones. In this paper, we will categorize data patterns taking into account degradation mechanisms and pattern dependency. In addition, we propose several modulation coding schemes to improve the data reliability by transforming original vulnerable data patterns into more robust ones.

cs.IT↗

arXiv2008

A splitting theorem for holomorphic Banach bundles

This paper is motivated by Grothendieck's splitting theorem. In the 1960s, Gohberg generalized this to a class of Banach bundles. We consider a compact complex manifold $X$ and a holomorphic Banach bundle $E \to X$ that is a compact perturbation of a trivial bundle in a sense recently introduced by Lempert. We prove that $E$ splits into the sum of a finite rank bundle and a trivial bundle, provided $H^{1}(X, Ø)=0$.

math.CV↗

arXiv2007

The Design of Efficiently-Encodable Rate-Compatible LDPC Codes

We present a new class of irregular low-density parity-check (LDPC) codes for moderate block lengths (up to a few thousand bits) that are well-suited for rate-compatible puncturing. The proposed codes show good performance under puncturing over a wide range of rates and are suitable for usage in incremental redundancy hybrid-automatic repeat request (ARQ) systems. In addition, these codes are linear-time encodable with simple shift-register circuits. For a block length of 1200 bits the codes outperform optimized irregular LDPC codes and extended irregular repeat-accumulate (eIRA) codes for all puncturing rates 0.6~0.9 (base code performance is almost the same) and are particularly good at high puncturing rates where good puncturing performance has been previously difficult to achieve.

cs.IT↗