PyTorchKR Papers

최신 논문

총 9,642편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1808.01477공식 코드 · Papers with Code

Learning Multi-scale Features for Foreground Segmentation

Foreground segmentation algorithms aim segmenting moving objects from the background in a robust way under various challenging scenarios. Encoder-decoder type deep neural networks that are used in this domain recently perform impressive segmentation results. In this work, we propose a novel robust encoder-decoder structure neural network that can be trained end-to-end using only a few training examples. The proposed method extends the Feature Pooling Module (FPM) of FgSegNet by introducing features fusions inside this module, which is capable of extracting multi-scale features within images; resulting in a robust feature pooling against camera motion, which can alleviate the need of multi-scale inputs to the network. Our method outperforms all existing state-of-the-art methods in CDnet2014 dataset by an average overall F-Measure of 0.9847. We also evaluate the effectiveness of our method on SBI2015 and UCSD Background Subtraction datasets. The source code of the proposed method is made available at https://github.com/lim-anggun/FgSegNet_v2 .

arXiv:1808.00892

Semi-blind source separation with multichannel variational autoencoder

This paper proposes a multichannel source separation technique called the multichannel variational autoencoder (MVAE) method, which uses a conditional VAE (CVAE) to model and estimate the power spectrograms of the sources in a mixture. By training the CVAE using the spectrograms of training examples with source-class labels, we can use the trained decoder distribution as a universal generative model capable of generating spectrograms conditioned on a specified class label. By treating the latent space variables and the class label as the unknown parameters of this generative model, we can develop a convergence-guaranteed semi-blind source separation algorithm that consists of iteratively estimating the power spectrograms of the underlying sources as well as the separation matrices. In experimental evaluations, our MVAE produced better separation performance than a baseline method.

arXiv:1808.00845

RGB Video Based Tennis Action Recognition Using a Deep Historical Long Short-Term Memory

Action recognition has attracted increasing attention from RGB input in computer vision partially due to potential applications on somatic simulation and statistics of sport such as virtual tennis game and tennis techniques and tactics analysis by video. Recently, deep learning based methods have achieved promising performance for action recognition. In this paper, we propose weighted Long Short-Term Memory adopted with convolutional neural network representations for three dimensional tennis shots recognition. First, the local two-dimensional convolutional neural network spatial representations are extracted from each video frame individually using a pre-trained Inception network. Then, a weighted Long Short-Term Memory decoder is introduced to take the output state at time t and the historical embedding feature at time t-1 to generate feature vector using a score weighting scheme. Finally, we use the adopted CNN and weighted LSTM to map the original visual features into a vector space to generate the spatial-temporal semantical description of visual sequences and classify the action video content. Experiments on the benchmark demonstrate that our method using only simple raw RGB video can achieve better performance than the state-of-the-art baselines for tennis shot recognition.

arXiv:1808.00671공식 코드 · Papers with Code

PCN: Point Completion Network

Shape completion, the problem of estimating the complete geometry of objects from partial observations, lies at the core of many vision and robotics applications. In this work, we propose Point Completion Network (PCN), a novel learning-based approach for shape completion. Unlike existing shape completion methods, PCN directly operates on raw point clouds without any structural assumption (e.g. symmetry) or annotation (e.g. semantic class) about the underlying shape. It features a decoder design that enables the generation of fine-grained completions while maintaining a small number of parameters. Our experiments show that PCN produces dense, complete point clouds with realistic structures in the missing regions on inputs with various levels of incompleteness and noise, including cars from LiDAR scans in the KITTI dataset.

arXiv:1808.00184

From Thumbnails to Summaries - A single Deep Neural Network to Rule Them All

Video summaries come in many forms, from traditional single-image thumbnails, animated thumbnails, storyboards, to trailer-like video summaries. Content creators use the summaries to display the most attractive portion of their videos; the users use them to quickly evaluate if a video is worth watching. All forms of summaries are essential to video viewers, content creators, and advertisers. Often video content management systems have to generate multiple versions of summaries that vary in duration and presentational forms. We present a framework ReconstSum that utilizes LSTM-based autoencoder architecture to extract and select a sparse subset of video frames or keyshots that optimally represent the input video in an unsupervised manner. The encoder selects a subset from the input video while the decoder seeks to reconstruct the video from the selection. The goal is to minimize the difference between the original input video and the reconstructed video. Our method is easily extendable to generate a variety of applications including static video thumbnails, animated thumbnails, storyboards and "trailer-like" highlights. We specifically study and evaluate two most popular use cases: thumbnail generation and storyboard generation. We demonstrate that our methods generate better results than the state-of-the-art techniques in both use cases.

arXiv:1807.11605

Doubly Attentive Transformer Machine Translation

In this paper a doubly attentive transformer machine translation model (DATNMT) is presented in which a doubly-attentive transformer decoder normally joins spatial visual features obtained via pretrained convolutional neural networks, conquering any gap between image captioning and translation. In this framework, the transformer decoder figures out how to take care of source-language words and parts of an image freely by methods for two separate attention components in an Enhanced Multi-Head Attention Layer of doubly attentive transformer, as it generates words in the target language. We find that the proposed model can effectively exploit not just the scarce multimodal machine translation data, but also large general-domain text-only machine translation corpora, or image-text image captioning corpora. The experimental results show that the proposed doubly-attentive transformer-decoder performs better than a single-decoder transformer model, and gives the state-of-the-art results in the English-German multimodal machine translation task.

arXiv:1807.10893

Back-Translation-Style Data Augmentation for End-to-End ASR

In this paper we propose a novel data augmentation method for attention-based end-to-end automatic speech recognition (E2E-ASR), utilizing a large amount of text which is not paired with speech signals. Inspired by the back-translation technique proposed in the field of machine translation, we build a neural text-to-encoder model which predicts a sequence of hidden states extracted by a pre-trained E2E-ASR encoder from a sequence of characters. By using hidden states as a target instead of acoustic features, it is possible to achieve faster attention learning and reduce computational cost, thanks to sub-sampling in E2E-ASR encoder, also the use of the hidden states can avoid to model speaker dependencies unlike acoustic features. After training, the text-to-encoder model generates the hidden states from a large amount of unpaired text, then E2E-ASR decoder is retrained using the generated hidden states as additional training data. Experimental evaluation using LibriSpeech dataset demonstrates that our proposed method achieves improvement of ASR performance and reduces the number of unknown words without the need for paired data.

arXiv:1807.10857

A Comparison of Techniques for Language Model Integration in Encoder-Decoder Speech Recognition

Attention-based recurrent neural encoder-decoder models present an elegant solution to the automatic speech recognition problem. This approach folds the acoustic model, pronunciation model, and language model into a single network and requires only a parallel corpus of speech and text for training. However, unlike in conventional approaches that combine separate acoustic and language models, it is not clear how to use additional (unpaired) text. While there has been previous work on methods addressing this problem, a thorough comparison among methods is still lacking. In this paper, we compare a suite of past methods and some of our own proposed methods for using unpaired text data to improve encoder-decoder models. For evaluation, we use the medium-sized Switchboard data set and the large-scale Google voice search and dictation data sets. Our results confirm the benefits of using unpaired text across a range of methods and data sets. Surprisingly, for first-pass decoding, the rather simple approach of shallow fusion performs best across data sets. However, for Google data sets we find that cold fusion has a lower oracle error rate and outperforms other approaches after second-pass rescoring on the Google voice search data set.

arXiv:1807.10831

MoCoNet: Motion Correction in 3D MPRAGE images using a Convolutional Neural Network approach

Purpose: The suppression of motion artefacts from MR images is a challenging task. The purpose of this paper is to develop a standalone novel technique to suppress motion artefacts from MR images using a data-driven deep learning approach. Methods: A deep learning convolutional neural network (CNN) was developed to remove motion artefacts in brain MR images. A CNN was trained on simulated motion corrupted images to identify and suppress artefacts due to the motion. The network was an encoder-decoder CNN architecture where the encoder decomposed the motion corrupted images into a set of feature maps. The feature maps were then combined by the decoder network to generate a motion-corrected image. The network was tested on an unseen simulated dataset and an experimental, motion corrupted in vivo brain dataset. Results: The trained network was able to suppress the motion artefacts in the simulated motion corrupted images, and the mean percentage error in the motion corrected images was 2.69 % with a standard deviation of 0.95 %. The network was able to effectively suppress the motion artefacts from the experimental dataset, demonstrating the generalisation capability of the trained network. Conclusion: A novel and generic motion correction technique has been developed that can suppress motion artefacts from motion corrupted MR images. The proposed technique is a standalone post-processing method that does not interfere with data acquisition or reconstruction parameters, thus making it suitable for a multitude of MR sequences.

arXiv:1807.11470

Deep Encoder-Decoder Models for Unsupervised Learning of Controllable Speech Synthesis

Generating versatile and appropriate synthetic speech requires control over the output expression separate from the spoken text. Important non-textual speech variation is seldom annotated, in which case output control must be learned in an unsupervised fashion. In this paper, we perform an in-depth study of methods for unsupervised learning of control in statistical speech synthesis. For example, we show that popular unsupervised training heuristics can be interpreted as variational inference in certain autoencoder models. We additionally connect these models to VQ-VAEs, another, recently-proposed class of deep variational autoencoders, which we show can be derived from a very similar mathematical argument. The implications of these new probabilistic interpretations are discussed. We illustrate the utility of the various approaches with an application to acoustic modelling for emotional speech synthesis, where the unsupervised methods for learning expression control (without access to emotional labels) are found to give results that in many aspects match or surpass the previous best supervised approach.

arXiv:1807.11433

REFUGE CHALLENGE 2018-Task 2:Deep Optic Disc and Cup Segmentation in Fundus Images Using U-Net and Multi-scale Feature Matching Networks

In this paper, an optic disc and cup segmentation method is proposed using U-Net followed by a multi-scale feature matching network. The proposed method targets task 2 of the REFUGE challenge 2018. In order to solve the segmentation problem of task 2, we firstly crop the input image using single shot multibox detector (SSD). The cropped image is then passed to an encoder-decoder network with skip connections also known as generator. Afterwards, both the ground truth and generated images are fed to a convolution neural network (CNN) to extract their multi-level features. A dice loss function is then used to match the features of the two images by minimizing the error at each layer. The aggregation of error from each layer is back-propagated through the generator network to enforce it to generate a segmented image closer to the ground truth. The CNN network improves the performance of the generator network without increasing the complexity of the model.

arXiv:1807.11407

HybridNet: Classification and Reconstruction Cooperation for Semi-Supervised Learning

In this paper, we introduce a new model for leveraging unlabeled data to improve generalization performances of image classifiers: a two-branch encoder-decoder architecture called HybridNet. The first branch receives supervision signal and is dedicated to the extraction of invariant class-related representations. The second branch is fully unsupervised and dedicated to model information discarded by the first branch to reconstruct input data. To further support the expected behavior of our model, we propose an original training objective. It favors stability in the discriminative branch and complementarity between the learned representations in the two branches. HybridNet is able to outperform state-of-the-art results on CIFAR-10, SVHN and STL-10 in various semi-supervised settings. In addition, visualizations and ablation studies validate our contributions and the behavior of the model on both CIFAR-10 and STL-10 datasets.

arXiv:1807.11234공식 코드 · Papers with Code

Improving Electron Micrograph Signal-to-Noise with an Atrous Convolutional Encoder-Decoder

We present an atrous convolutional encoder-decoder trained to denoise 512$\times$512 crops from electron micrographs. It consists of a modified Xception backbone, atrous convoltional spatial pyramid pooling module and a multi-stage decoder. Our neural network was trained end-to-end to remove Poisson noise applied to low-dose ($\ll$ 300 counts ppx) micrographs created from a new dataset of 17267 2048$\times$2048 high-dose ($>$ 2500 counts ppx) micrographs and then fine-tuned for ordinary doses (200-2500 counts ppx). Its performance is benchmarked against bilateral, non-local means, total variation, wavelet, Wiener and other restoration methods with their default parameters. Our network outperforms their best mean squared error and structural similarity index performances by 24.6% and 9.6% for low doses and by 43.7% and 5.5% for ordinary doses. In both cases, our network's mean squared error has the lowest variance. Source code and links to our new high-quality dataset and trained network have been made publicly available at https://github.com/Jeffrey-Ede/Electron-Micrograph-Denoiser

arXiv:1807.11219

Training Neural Machine Translation using Word Embedding-based Loss

In neural machine translation (NMT), the computational cost at the output layer increases with the size of the target-side vocabulary. Using a limited-size vocabulary instead may cause a significant decrease in translation quality. This trade-off is derived from a softmax-based loss function that handles in-dictionary words independently, in which word similarity is not considered. In this paper, we propose a novel NMT loss function that includes word similarity in forms of distances in a word embedding space. The proposed loss function encourages an NMT decoder to generate words close to their references in the embedding space; this helps the decoder to choose similar acceptable words when the actual best candidates are not included in the vocabulary due to its size limitation. In experiments using ASPEC Japanese-to-English and IWSLT17 English-to-French data sets, the proposed method showed improvements against a standard NMT baseline in both datasets; especially with IWSLT17 En-Fr, it achieved up to +1.72 in BLEU and +1.99 in METEOR. When the target-side vocabulary was very limited to 1,000 words, the proposed method demonstrated a substantial gain, +1.72 in METEOR with ASPEC Ja-En.

arXiv:1807.11079공식 코드 · Papers with Code

ReenactGAN: Learning to Reenact Faces via Boundary Transfer

We present a novel learning-based framework for face reenactment. The proposed method, known as ReenactGAN, is capable of transferring facial movements and expressions from monocular video input of an arbitrary person to a target person. Instead of performing a direct transfer in the pixel space, which could result in structural artifacts, we first map the source face onto a boundary latent space. A transformer is subsequently used to adapt the boundary of source face to the boundary of target face. Finally, a target-specific decoder is used to generate the reenacted target face. Thanks to the effective and reliable boundary-based transfer, our method can perform photo-realistic face reenactment. In addition, ReenactGAN is appealing in that the whole reenactment process is purely feed-forward, and thus the reenactment process can run in real-time (30 FPS on one GTX 1080 GPU). Dataset and model will be publicly available at https://wywu.github.io/projects/ReenactGAN/ReenactGAN.html

arXiv:1807.10547

CrossNet: An End-to-end Reference-based Super Resolution Network using Cross-scale Warping

The Reference-based Super-resolution (RefSR) super-resolves a low-resolution (LR) image given an external high-resolution (HR) reference image, where the reference image and LR image share similar viewpoint but with significant resolution gap x8. Existing RefSR methods work in a cascaded way such as patch matching followed by synthesis pipeline with two independently defined objective functions, leading to the inter-patch misalignment, grid effect and inefficient optimization. To resolve these issues, we present CrossNet, an end-to-end and fully-convolutional deep neural network using cross-scale warping. Our network contains image encoders, cross-scale warping layers, and fusion decoder: the encoder serves to extract multi-scale features from both the LR and the reference images; the cross-scale warping layers spatially aligns the reference feature map with the LR feature map; the decoder finally aggregates feature maps from both domains to synthesize the HR output. Using cross-scale warping, our network is able to perform spatial alignment at pixel-level in an end-to-end fashion, which improves the existing schemes both in precision (around 2dB-4dB) and efficiency (more than 100 times faster).

arXiv:1807.10299

Variational Option Discovery Algorithms

We explore methods for option discovery based on variational inference and make two algorithmic contributions. First: we highlight a tight connection between variational option discovery methods and variational autoencoders, and introduce Variational Autoencoding Learning of Options by Reinforcement (VALOR), a new method derived from the connection. In VALOR, the policy encodes contexts from a noise distribution into trajectories, and the decoder recovers the contexts from the complete trajectories. Second: we propose a curriculum learning approach where the number of contexts seen by the agent increases whenever the agent's performance is strong enough (as measured by the decoder) on the current set of contexts. We show that this simple trick stabilizes training for VALOR and prior variational option discovery methods, allowing a single agent to learn many more modes of behavior than it could with a fixed context distribution. Finally, we investigate other topics related to variational option discovery, including fundamental limitations of the general approach and the applicability of learned options to downstream tasks.

arXiv:1807.09950공식 코드 · Papers with Code

Variational Memory Encoder-Decoder

Introducing variability while maintaining coherence is a core task in learning to generate utterances in conversation. Standard neural encoder-decoder models and their extensions using conditional variational autoencoder often result in either trivial or digressive responses. To overcome this, we explore a novel approach that injects variability into neural encoder-decoder via the use of external memory as a mixture model, namely Variational Memory Encoder-Decoder (VMED). By associating each memory read with a mode in the latent mixture distribution at each timestep, our model can capture the variability observed in sequential data such as natural conversations. We empirically compare the proposed model against other recent approaches on various conversational datasets. The results show that VMED consistently achieves significant improvement over others in both metric-based and qualitative evaluations.

arXiv:1807.09937

HiDDeN: Hiding Data With Deep Networks

Recent work has shown that deep neural networks are highly sensitive to tiny perturbations of input images, giving rise to adversarial examples. Though this property is usually considered a weakness of learned models, we explore whether it can be beneficial. We find that neural networks can learn to use invisible perturbations to encode a rich amount of useful information. In fact, one can exploit this capability for the task of data hiding. We jointly train encoder and decoder networks, where given an input message and cover image, the encoder produces a visually indistinguishable encoded image, from which the decoder can recover the original message. We show that these encodings are competitive with existing data hiding algorithms, and further that they can be made robust to noise: our models learn to reconstruct hidden information in an encoded image despite the presence of Gaussian blurring, pixel-wise dropout, cropping, and JPEG compression. Even though JPEG is non-differentiable, we show that a robust model can be trained using differentiable approximations. Finally, we demonstrate that adversarial training improves the visual quality of encoded images.

arXiv:1807.10201

A Style-Aware Content Loss for Real-time HD Style Transfer

Recently, style transfer has received a lot of attention. While much of this research has aimed at speeding up processing, the approaches are still lacking from a principled, art historical standpoint: a style is more than just a single image or an artist, but previous work is limited to only a single instance of a style or shows no benefit from more images. Moreover, previous work has relied on a direct comparison of art in the domain of RGB images or on CNNs pre-trained on ImageNet, which requires millions of labeled object bounding boxes and can introduce an extra bias, since it has been assembled without artistic consideration. To circumvent these issues, we propose a style-aware content loss, which is trained jointly with a deep encoder-decoder network for real-time, high-resolution stylization of images and videos. We propose a quantitative measure for evaluating the quality of a stylized image and also have art historians rank patches from our approach against those from previous work. These and our qualitative results ranging from small image patches to megapixel stylistic images and videos show that our approach better captures the subtle nature in which a style affects content.