No abstract

This paper deals with acoustic event detection (AED), such as screams, gunshots and explosions, in noisy environments. The main aim is to improve the detection performance under adverse conditions with a very low signal-to-noise ratio (SNR). A novel filtering method combined with an energy detector is presented. The wavelet packet transform (WPT) is first used for time-frequency representation of the acoustic signal. The proposed filter in the wavelet packet domain then uses a priori knowledge of the target event and an estimate of noise features to selectively suppress the background noise. It is in fact a content-aware band-pass filter which can automatically pass the frequency bands that are more significant in the target than in the noise. Theoretical analysis shows that the proposed filtering method is capable of enhancing the target content while suppressing the background noise, for signals with a low SNR. A condition to increase the probability of correct detection is also obtained. Experiments have been carried out on a large dataset of acoustic events that are contaminated by different types of environmental noise and white noise with varying SNRs. Results show that the proposed method is more robust and better adapted to noise than ordinary energy detectors and it can work even with an SNR as low as -15dB. A practical system for real time processing and multi-target detection is also proposed in this work.

State-of-the-art techniques in Generative Adversarial Networks (GANs) have shown remarkable success in image-to-image translation from peer domain X to domain Y using paired image data. However, obtaining abundant paired data is a non-trivial and expensive process in the majority of applications. When there is a need to translate images across n domains, if the training is performed between every two domains, the complexity of the training will increase quadratically. Moreover, training with data from two domains only at a time cannot benefit from data of other domains, which prevents the extraction of more useful features and hinders the progress of this research area. In this work, we propose a general framework for unsupervised image-to-image translation across multiple domains, which can translate images from domain X to any a domain without requiring direct training between the two domains involved in image translation. A byproduct of the framework is the reduction of computing time and computing resources, since it needs less time than training the domains in pairs as is done in state-of-the-art works. Our proposed framework consists of a pair of encoders along with a pair of GANs which learns high-level features across different domains to generate diverse and realistic samples from. Our framework shows competing results on many image-to-image tasks compared with state-of-the-art techniques. CCS CONCEPTS • Computing methodologies → Image representations; Neural networks; Unsupervised learning;

Generating accurate descriptions for online fashion items is important not only for enhancing customers' shopping experiences, but also for the increase of online sales. Besides the need of correctly presenting the attributes of items, the expressions in an enchanting style could better attract customer interests. The goal of this work is to develop a novel learning framework for accurate and expressive fashion captioning. Different from popular work on image captioning, it is hard to identify and describe the rich attributes of fashion items. We seed the description of an item by first identifying its attributes, and introduce attribute-level semantic (ALS) reward and sentence-level semantic (SLS) reward as metrics to improve the quality of text descriptions. We further integrate the training of our model with maximum likelihood estimation (MLE), attribute embedding, and Reinforcement Learning (RL). To facilitate the learning, we build a new FAshion CAptioning Dataset (FACAD), which contains 993K images and 130K corresponding enchanting and diverse descriptions. Experiments on FACAD demonstrate the effectiveness of our model.

Learning target side syntactic structure has been shown to improve Neural Machine Translation (NMT). However, incorporating syntax through latent variables introduces additional complexity in inference, as the models need to marginalize over the latent syntactic structures. To avoid this, models often resort to greedy search which only allows them to explore a limited portion of the latent space.