Conditional Hyper-Network for Blind Super-Resolution With Multiple Degradations.

Although the single-image super-resolution (SISR) methods have achieved great success on the single degradation, they still suffer performance drop with multiple degrading effects in real scenarios. Recently, some blind and non-blind models for multiple degradations have been explored. However, these methods usually degrade significantly for distribution shifts between the training and test data. Towards this end, we propose a novel conditional hyper-network framework for super-resolution with multiple degradations (named CMDSR), which helps the SR framework learn how to adapt to changes in the degradation distribution of input. We extract degradation prior at the task-level with the proposed ConditionNet, which will be used to adapt the parameters of the basic SR network (BaseNet). Specifically, the ConditionNet of our framework first learns the degradation prior from a support set, which is composed of a series of degraded image patches from the same task. Then the adaptive BaseNet rapidly shifts its parameters according to the conditional features. Moreover, in order to better extract degradation prior, we propose a task contrastive loss to shorten the inner-task distance and enlarge the cross-task distance between task-level features. Without predefining degradation maps, our blind framework can conduct one single parameter update to yield considerable improvement in SR results. Extensive experiments demonstrate the effectiveness of CMDSR over various blind, and even several non-blind methods. The flexible BaseNet structure also reveals that CMDSR can be a general framework for a large series of SISR models. Our code is available at https://github.com/guanghaoyin/CMDSR.

And the nominees are: Using design-awards datasets to build computational aesthetic evaluation model.

Aesthetic perception is a human instinct that is responsive to multimedia stimuli. Giving computers the ability to assess human sensory and perceptual experience of aesthetics is a well-recognized need for the intelligent design industry and multimedia intelligence study. In this work, we constructed a novel database for the aesthetic evaluation of design, using 2,918 images collected from the archives of two major design awards, and we also present a method of aesthetic evaluation that uses machine learning algorithms. Reviewers' ratings of the design works are set as the ground-truth annotations for the dataset. Furthermore, multiple image features are extracted and fused. The experimental results demonstrate the validity of the proposed approach. Primary screening using aesthetic computing can be an intelligent assistant for various design evaluations and can reduce misjudgment in art and design review due to visual aesthetic fatigue after a long period of viewing. The study of computational aesthetic evaluation can provide positive effect on the efficiency of design review, and it is of great significance to aesthetic recognition exploration and applications development.

A Low-Cost End-to-End sEMG-Based Gait Sub-Phase Recognition System.

As surface electromyogram (sEMG) signals have the ability to detect human movement intention, they are commonly used to be control inputs. However, gait sub-phase classification typically requires monotonous manual labeling process, and commercial sEMG acquisition devices are quite bulky and expensive, thus current sEMG-based gait sub-phase recognition systems are complex and have poor portability. This study presents a low-cost but effective end-to-end sEMG-based gait sub-phase recognition system, which contains a wireless multi-channel signal acquisition device simultaneously collecting sEMG of thigh muscles and plantar pressure signals, and a novel neural network-based sEMG signal classifier combining long-short term memory (LSTM) with multilayer perceptron (MLP). We evaluated the system with subjects walking under five conditions: flat terrain at 5 km/h, flat terrain at 3 km/h, 20 kg backpack at 5 km/h, 20 kg shoulder bag at 5 km/h and 15° slope at 5 km/h. Experimental results show that the proposed method achieved average classification accuracies of 94.10%, 87.25%, 90.71%, 94.02%, and 87.87%, respectively, which were significantly higher than existing recognition methods. Additionally, the proposed system had a good real-time performance with low average inference time in the range of 3.25 ~ 3.31 ms.

A Brain-Machine Interface Based on ERD/ERS for an Upper-Limb Exoskeleton Control.

To recognize the user's motion intention, brain-machine interfaces (BMI) usually decode movements from cortical activity to control exoskeletons and neuroprostheses for daily activities. The aim of this paper is to investigate whether self-induced variations of the electroencephalogram (EEG) can be useful as control signals for an upper-limb exoskeleton developed by us. A BMI based on event-related desynchronization/synchronization (ERD/ERS) is proposed. In the decoder-training phase, we investigate the offline classification performance of left versus right hand and left hand versus both feet by using motor execution (ME) or motor imagery (MI). The results indicate that the accuracies of ME sessions are higher than those of MI sessions, and left hand versus both feet paradigm achieves a better classification performance, which would be used in the online-control phase. In the online-control phase, the trained decoder is tested in two scenarios (wearing or without wearing the exoskeleton). The MI and ME sessions wearing the exoskeleton achieve mean classification accuracy of 84.29% ± 2.11% and 87.37% ± 3.06%, respectively. The present study demonstrates that the proposed BMI is effective to control the upper-limb exoskeleton, and provides a practical method by non-invasive EEG signal associated with human natural behavior for clinical applications.

An upper-limb power-assist exoskeleton using proportional myoelectric control.

We developed an upper-limb power-assist exoskeleton actuated by pneumatic muscles. The exoskeleton included two metal links: a nylon joint, four size-adjustable carbon fiber bracers, a potentiometer and two pneumatic muscles. The proportional myoelectric control method was proposed to control the exoskeleton according to the user's motion intention in real time. With the feature extraction procedure and the classification (back-propagation neural network), an electromyogram (EMG)-angle model was constructed to be used for pattern recognition. Six healthy subjects performed elbow flexion-extension movements under four experimental conditions: (1) holding a 1-kg load, wearing the exoskeleton, but with no actuation and for different periods (2-s, 4-s and 8-s periods); (2) holding a 1-kg load, without wearing the exoskeleton, for a fixed period; (3) holding a 1-kg load, wearing the exoskeleton, but with no actuation, for a fixed period; (4) holding a 1-kg load, wearing the exoskeleton under proportional myoelectric control, for a fixed period. The EMG signals of the biceps brachii, the brachioradialis, the triceps brachii and the anconeus and the angle of the elbow were collected. The control scheme's reliability and power-assist effectiveness were evaluated in the experiments. The results indicated that the exoskeleton could be controlled by the user's motion intention in real time and that it was useful for augmenting arm performance with neurological signal control, which could be applied to assist in elbow rehabilitation after neurological injury.