The central building block of convolutional neural networks (CNNs) is the convolution operator, which enables networks to construct informative features by fusing both spatial and channel-wise information within local receptive fields at each layer. A broad range of prior research has investigated the spatial component of this relationship, seeking to strengthen the representational power of a CNN by enhancing the quality of spatial encodings throughout its feature hierarchy. In this work, we focus instead on the channel relationship and propose a novel architectural unit, which we term the “Squeeze-and-Excitation” (SE) block, that adaptively recalibrates channel-wise feature responses by explicitly modelling interdependencies between channels. We show that these blocks can be stacked together to form SENet architectures that generalise extremely effectively across different datasets. We further demonstrate that SE blocks bring significant improvements in performance for existing state-of-the-art CNNs at slight additional computational cost. Squeeze-and-Excitation Networks formed the foundation of our ILSVRC 2017 classification submission which won first place and reduced the top-5 error to 2.251 percent, surpassing the winning entry of 2016 by a relative improvement of ${\sim }$ ∼ 25 percent. Models and code are available at https://github.com/hujie-frank/SENet .

/pdf/squeeze-and-excitation-networks-50icgwqgeu.pdf

Squeeze-and-Excitation Networks

Preparing the books to read every day is enjoyable for many people. However, there are still many people who also don't like reading. This is a problem. But, when you can support others to start reading, it will be better. One of the books that can be recommended for new readers is experience and education. This book is not kind of difficult book to read. It can be read and understand by the new readers.

Experience and Education.

The central building block of convolutional neural networks (CNNs) is the convolution operator, which enables networks to construct informative features by fusing both spatial and channel-wise information within local receptive fields at each layer. A broad range of prior research has investigated the spatial component of this relationship, seeking to strengthen the representational power of a CNN by enhancing the quality of spatial encodings throughout its feature hierarchy. In this work, we focus instead on the channel relationship and propose a novel architectural unit, which we term the "Squeeze-and-Excitation" (SE) block, that adaptively recalibrates channel-wise feature responses by explicitly modelling interdependencies between channels. We show that these blocks can be stacked together to form SENet architectures that generalise extremely effectively across different datasets. We further demonstrate that SE blocks bring significant improvements in performance for existing state-of-the-art CNNs at slight additional computational cost. Squeeze-and-Excitation Networks formed the foundation of our ILSVRC 2017 classification submission which won first place and reduced the top-5 error to 2.251%, surpassing the winning entry of 2016 by a relative improvement of ~25%. Models and code are available at this https URL.

Point clouds provide a flexible geometric representation suitable for countless applications in computer graphics; they also comprise the raw output of most 3D data acquisition devices. While hand-designed features on point clouds have long been proposed in graphics and vision, however, the recent overwhelming success of convolutional neural networks (CNNs) for image analysis suggests the value of adapting insight from CNN to the point cloud world. Point clouds inherently lack topological information, so designing a model to recover topology can enrich the representation power of point clouds. To this end, we propose a new neural network module dubbed EdgeConv suitable for CNN-based high-level tasks on point clouds, including classification and segmentation. EdgeConv acts on graphs dynamically computed in each layer of the network. It is differentiable and can be plugged into existing architectures. Compared to existing modules operating in extrinsic space or treating each point independently, EdgeConv has several appealing properties: It incorporates local neighborhood information; it can be stacked applied to learn global shape properties; and in multi-layer systems affinity in feature space captures semantic characteristics over potentially long distances in the original embedding. We show the performance of our model on standard benchmarks, including ModelNet40, ShapeNetPart, and S3DIS.

/pdf/dynamic-graph-cnn-for-learning-on-point-clouds-20wlcru9tg.pdf

Dynamic Graph CNN for Learning on Point Clouds

Synthesizing high-quality images from text descriptions is a challenging problem in computer vision and has many practical applications. Samples generated by existing textto- image approaches can roughly reflect the meaning of the given descriptions, but they fail to contain necessary details and vivid object parts. In this paper, we propose Stacked Generative Adversarial Networks (StackGAN) to generate 256.256 photo-realistic images conditioned on text descriptions. We decompose the hard problem into more manageable sub-problems through a sketch-refinement process. The Stage-I GAN sketches the primitive shape and colors of the object based on the given text description, yielding Stage-I low-resolution images. The Stage-II GAN takes Stage-I results and text descriptions as inputs, and generates high-resolution images with photo-realistic details. It is able to rectify defects in Stage-I results and add compelling details with the refinement process. To improve the diversity of the synthesized images and stabilize the training of the conditional-GAN, we introduce a novel Conditioning Augmentation technique that encourages smoothness in the latent conditioning manifold. Extensive experiments and comparisons with state-of-the-arts on benchmark datasets demonstrate that the proposed method achieves significant improvements on generating photo-realistic images conditioned on text descriptions.

StackGAN: Text to Photo-Realistic Image Synthesis with Stacked Generative Adversarial Networks

Continuing interest in digital games indicated that it would be useful to update Connolly et al.'s (2012) systematic literature review of empirical evidence about the positive impacts and outcomes of games. Since a large number of papers was identified in the period from 2009 to 2014, the current review focused on 143 papers that provided higher quality evidence about the positive outcomes of games. Connolly et al.'s multidimensional analysis of games and their outcomes provided a useful framework for organising the varied research in this area. The most frequently occurring outcome reported for games for learning was knowledge acquisition, while entertainment games addressed a broader range of affective, behaviour change, perceptual and cognitive and physiological outcomes. Games for learning were found across varied topics with STEM subjects and health the most popular. Future research on digital games would benefit from a systematic programme of experimental work, examining in detail which game features are most effective in promoting engagement and supporting learning. The current systematic literature review updates Author (date).The review looks at impacts and outcomes of playing digital games from 2009 to 2014.Multi-component coding of papers, games and learning outcomes was used.Many papers were found with 143 papers providing high quality evidence.Games for entertainment and learning addressed different outcomes.

/pdf/an-update-to-the-systematic-literature-review-of-empirical-oy90nea5x6.pdf

An update to the systematic literature review of empirical evidence of the impacts and outcomes of computer games and serious games

Although there is a consensus on the instructional potential of Serious Games (SGs), there is still a lack of methodologies and tools not only for design but also to support analysis and assessment. Filling this gap is one of the main aims of the Games and Learning Alliance (http://www.galanoe.eu) European Network of Excellence on Serious Games, which has a focus upon pedagogy-driven SGs. This paper relies on the assumption that the fundamental aspect of SG design consists in the translation of learning goals/practices into mechanical element of gameplay, serving to an instructional purpose beside that of play and fun. This paper proposes the Learning Mechanics–Game Mechanics (LM-GM) model, which supports SG analysis and design by allowing reflection on the various pedagogical and game elements in an SG. The LM-GM model includes a set of pre-defined game mechanics and pedagogical elements that we have abstracted from literature on game studies and learning theories.

Designers and analysts can exploit these mechanics to draw the LM-GM map for a game, so as to identify and highlight its main pedagogical and entertainment features, and their interrelations. The tool may also be useful for teachers to evaluate the effectiveness of a given game and better understand how to implement it in educational settings. A case study is reported to illustrate the framework's support in determining how gameplay and pedagogy intertwine in an SG. Finally, the paper presents the results of two comparative user tests demonstrating the advantages of the proposed model with respect to a similar state-of-the-art framework.

/pdf/mapping-learning-and-game-mechanics-for-serious-games-nevwa2jsvn.pdf

Mapping learning and game mechanics for serious games analysis

Designing architectures for deep neural networks requires expert knowledge and substantial computation time. We propose a technique to accelerate architecture selection by learning an auxiliary HyperNet that generates the weights of a main model conditioned on that model's architecture. By comparing the relative validation performance of networks with HyperNet-generated weights, we can effectively search over a wide range of architectures at the cost of a single training run. To facilitate this search, we develop a flexible mechanism based on memory read-writes that allows us to define a wide range of network connectivity patterns, with ResNet, DenseNet, and FractalNet blocks as special cases. We validate our method (SMASH) on CIFAR-10 and CIFAR-100, STL-10, ModelNet10, and Imagenet32x32, achieving competitive performance with similarly-sized hand-designed networks. Our code is available at this https URL

/pdf/smash-one-shot-model-architecture-search-through-31dxypjgzt.pdf

SMASH: One-Shot Model Architecture Search through HyperNetworks

When working with three-dimensional data, choice of representation is key. We explore voxel-based models, and present evidence for the viability of voxellated representations in applications including shape modeling and object classification. Our key contributions are methods for training voxel-based variational autoencoders, a user interface for exploring the latent space learned by the autoencoder, and a deep convolutional neural network architecture for object classification. We address challenges unique to voxel-based representations, and empirically evaluate our models on the ModelNet benchmark, where we demonstrate a 51.5% relative improvement in the state of the art for object classification.

Generative and Discriminative Voxel Modeling with Convolutional Neural Networks

The increasingly photorealistic sample quality of generative image models suggests their feasibility in applications beyond image generation. We present the Neural Photo Editor, an interface that leverages the power of generative neural networks to make large, semantically coherent changes to existing images. To tackle the challenge of achieving accurate reconstructions without loss of feature quality, we introduce the Introspective Adversarial Network, a novel hybridization of the VAE and GAN. Our model efficiently captures long-range dependencies through use of a computational block based on weight-shared dilated convolutions, and improves generalization performance with Orthogonal Regularization, a novel weight regularization method. We validate our contributions on CelebA, SVHN, and CIFAR-100, and produce samples and reconstructions with high visual fidelity.

/pdf/neural-photo-editing-with-introspective-adversarial-networks-2ws8gdol1y.pdf

Theodore Lim

Papers

An update to the systematic literature review of empirical evidence of the impacts and outcomes of computer games and serious games

Mapping learning and game mechanics for serious games analysis

SMASH: One-Shot Model Architecture Search through HyperNetworks

Generative and Discriminative Voxel Modeling with Convolutional Neural Networks

Neural photo editing with introspective adversarial networks