Emotion Transfer for 3D Hand and Full Body Motion Using StarGAN

Abstract

In this paper, we propose a new data-driven framework for 3D hand and full-body motion emotion transfer. Specifically, we formulate the motion synthesis task as an image-to-image translation problem. By presenting a motion sequence as an image representation, the emotion can be transferred by our framework using StarGAN. To evaluate our proposed method’s effectiveness, we first conducted a user study to validate the perceived emotion from the captured and synthesized hand motions. We further evaluate the synthesized hand and full body motions qualitatively and quantitatively. Experimental results show that our synthesized motions are comparable to the captured motions and those created by an existing method in terms of naturalness and visual quality.

Internet and Biometric Web Based Business Management Decision Support

“… Among clustering algorithms the most common choices in biometrics or neuroscience research are: K-Means Clustering [185,186,187], Fuzzy C-means Algorithm [188,189], Expectation-Maximization (EM) Algorithm [190], and Hierarchical Clustering Algorithm [188,191,192].  Among reinforcement learning algorithms the most common choices are: deep reinforcement learning [193,194,195] and inverse reinforcement learning [196].…”

A Comprehensive Survey on Artificial Intelligence for Unmanned Aerial Vehicles

“…The KNN algorithm is used to classify human actions and estimate the full-body position of the human operator. The swarm bases its objective direction on the Policy-based Reinforcement Learning [57], [58] Address the limitations of PID control Value-based Reinforcement Learning [59], [60] Long-term resource allocation [61] Inefficiency of existing route planning algorithms [62] UAV navigation and obstacle avoidance Imitation RL learning [63] To detect the precise motions necessary to lead the UAV along the trajectories [64] Enhancing UAV tracking performance [65] UAV deployment strategy to maximize UAV owner profit and on-ground user benefits Inverse reinforcement learning [117], [67] To track a multirotor UAV's path Model-ensemble based Hybrid RL algorithms [68] To predict wheat output using UAVs in the winter [69] To evaluate total nitrogen concentration in water [70] Coordination among several UAVs traveling over a vast region LeNet Shallow Convolutional Neural Networks [72], [73] For spreading deep neural networks (DNNs) within unmanned aerial vehicles and to adjust the system to the UAV's dynamic movement and network fluctuation (UAVs) [118] Incorporating machine learning (ML) capabilities into small UAVs AlexNet Shallow Convolutional Neural Networks [80] To automatically detect damage to wind turbine blade surfaces Deep Residual Learning Network (ResNet) [83] For real-time UAV identification InceptionNet [85] To aid UAV-based surveillance operations that include the collection of movies using a mobile camera Deep Convolutional GANs (DCGANs) [92] To enable 5G-enabled maritime UAV communication employing millimeter wave (mmWave) for the air-to-surface link Wasserstein GANs (WGANs) [95] To enhance wireless signal-based detection of unauthorized UAVs StarGANs [119] Transmitting emotions using 3D hand and full-body motion CycleGANs [98] To reduce wildfire damage Long Short-Term Memory (LSTM) in recurrent neural networks (RNNs) [104] Resource allocation issue for UAVs [105] UAV anomaly detection [106] UAV communication for future wireless networks Gated Rec...…”

A Review of AI Cloud and Edge Sensors, Methods, and Applications for the Recognition of Emotional, Affective and Physiological States

“… Among clustering algorithms the most common choices in biometrics or neuroscience research are: K-Means Clustering [ 185 , 186 , 187 ], Fuzzy C-means Algorithm [ 188 , 189 ], Expectation-Maximization (EM) Algorithm [ 190 ], and Hierarchical Clustering Algorithm [ 188 , 191 , 192 ]. Among reinforcement learning algorithms the most common choices are: deep reinforcement learning [ 193 , 194 , 195 ] and inverse reinforcement learning [ 196 ]. …”

“…Among reinforcement learning algorithms the most common choices are: deep reinforcement learning [ 193 , 194 , 195 ] and inverse reinforcement learning [ 196 ].…”