arXiv subjects

Anh Nguyen

Publications and source records attributed to Anh Nguyen.

At least 217 records · Page 12Linked to original sources

arXiv2019

V2CNet: A Deep Learning Framework to Translate Videos to Commands for Robotic Manipulation

We propose V2CNet, a new deep learning framework to automatically translate the demonstration videos to commands that can be directly used in robotic applications. Our V2CNet has two branches and aims at understanding the demonstration video in a fine-grained manner. The first branch has the encoder-decoder architecture to encode the visual features and sequentially generate the output words as a command, while the second branch uses a Temporal Convolutional Network (TCN) to learn the fine-grained actions. By jointly training both branches, the network is able to model the sequential information of the command, while effectively encodes the fine-grained actions. The experimental results on our new large-scale dataset show that V2CNet outperforms recent state-of-the-art methods by a substantial margin, while its output can be applied in real robotic applications. The source code and trained models will be made available.

cs.CV↗

arXiv2018

VectorDefense: Vectorization as a Defense to Adversarial Examples

Training deep neural networks on images represented as grids of pixels has brought to light an interesting phenomenon known as adversarial examples. Inspired by how humans reconstruct abstract concepts, we attempt to codify the input bitmap image into a set of compact, interpretable elements to avoid being fooled by the adversarial structures. We take the first step in this direction by experimenting with image vectorization as an input transformation step to map the adversarial examples back into the natural manifold of MNIST handwritten digits. We compare our method vs. state-of-the-art input transformations and further discuss the trade-offs between a hand-designed and a learned transformation defense.

cs.CV↗

arXiv2018

Real-Time 6DOF Pose Relocalization for Event Cameras with Stacked Spatial LSTM Networks

We present a new method to relocalize the 6DOF pose of an event camera solely based on the event stream. Our method first creates the event image from a list of events that occurs in a very short time interval, then a Stacked Spatial LSTM Network (SP-LSTM) is used to learn the camera pose. Our SP-LSTM is composed of a CNN to learn deep features from the event images and a stack of LSTM to learn spatial dependencies in the image feature space. We show that the spatial dependency plays an important role in the relocalization task and the SP-LSTM can effectively learn this information. The experimental results on a publicly available dataset show that our approach generalizes well and outperforms recent methods by a substantial margin. Overall, our proposed method reduces by approx. 6 times the position error and 3 times the orientation error compared to the current state of the art. The source code and trained models will be released.

cs.CV↗

arXiv2018

Object Captioning and Retrieval with Natural Language

We address the problem of jointly learning vision and language to understand the object in a fine-grained manner. The key idea of our approach is the use of object descriptions to provide the detailed understanding of an object. Based on this idea, we propose two new architectures to solve two related problems: object captioning and natural language-based object retrieval. The goal of the object captioning task is to simultaneously detect the object and generate its associated description, while in the object retrieval task, the goal is to localize an object given an input query. We demonstrate that both problems can be solved effectively using hybrid end-to-end CNN-LSTM networks. The experimental results on our new challenging dataset show that our methods outperform recent methods by a fair margin, while providing a detailed understanding of the object and having fast inference time. The source code will be made available.

cs.CV↗

arXiv2018

AffordanceNet: An End-to-End Deep Learning Approach for Object Affordance Detection

We propose AffordanceNet, a new deep learning approach to simultaneously detect multiple objects and their affordances from RGB images. Our AffordanceNet has two branches: an object detection branch to localize and classify the object, and an affordance detection branch to assign each pixel in the object to its most probable affordance label. The proposed framework employs three key components for effectively handling the multiclass problem in the affordance mask: a sequence of deconvolutional layers, a robust resizing strategy, and a multi-task loss function. The experimental results on the public datasets show that our AffordanceNet outperforms recent state-of-the-art methods by a fair margin, while its end-to-end architecture allows the inference at the speed of 150ms per image. This makes our AffordanceNet well suitable for real-time robotic applications. Furthermore, we demonstrate the effectiveness of AffordanceNet in different testing environments and in real robotic applications. The source code is available at https://github.com/nqanh/affordance-net

cs.CV↗

arXiv2017

Spatial PixelCNN: Generating Images from Patches

In this paper we propose Spatial PixelCNN, a conditional autoregressive model that generates images from small patches. By conditioning on a grid of pixel coordinates and global features extracted from a Variational Autoencoder (VAE), we are able to train on patches of images, and reproduce the full-sized image. We show that it not only allows for generating high quality samples at the same resolution as the underlying dataset, but is also capable of upscaling images to arbitrary resolutions (tested at resolutions up to $50\times$) on the MNIST dataset. Compared to a PixelCNN++ baseline, Spatial PixelCNN quantitatively and qualitatively achieves similar performance on the MNIST dataset.

cs.CV↗

arXiv2017

Automatically identifying, counting, and describing wild animals in camera-trap images with deep learning

Having accurate, detailed, and up-to-date information about the location and behavior of animals in the wild would revolutionize our ability to study and conserve ecosystems. We investigate the ability to automatically, accurately, and inexpensively collect such data, which could transform many fields of biology, ecology, and zoology into "big data" sciences. Motion sensor "camera traps" enable collecting wildlife pictures inexpensively, unobtrusively, and frequently. However, extracting information from these pictures remains an expensive, time-consuming, manual task. We demonstrate that such information can be automatically extracted by deep learning, a cutting-edge type of artificial intelligence. We train deep convolutional neural networks to identify, count, and describe the behaviors of 48 species in the 3.2-million-image Snapshot Serengeti dataset. Our deep neural networks automatically identify animals with over 93.8% accuracy, and we expect that number to improve rapidly in years to come. More importantly, if our system classifies only images it is confident about, our system can automate animal identification for 99.3% of the data while still performing at the same 96.6% accuracy as that of crowdsourced teams of human volunteers, saving more than 8.4 years (at 40 hours per week) of human labeling effort (i.e. over 17,000 hours) on this 3.2-million-image dataset. Those efficiency gains immediately highlight the importance of using deep neural networks to automate data extraction from camera-trap images. Our results suggest that this technology could enable the inexpensive, unobtrusive, high-volume, and even real-time collection of a wealth of information about vast numbers of animals in the wild.

cs.CV↗

arXiv2017

Translating Videos to Commands for Robotic Manipulation with Deep Recurrent Neural Networks

We present a new method to translate videos to commands for robotic manipulation using Deep Recurrent Neural Networks (RNN). Our framework first extracts deep features from the input video frames with a deep Convolutional Neural Networks (CNN). Two RNN layers with an encoder-decoder architecture are then used to encode the visual features and sequentially generate the output words as the command. We demonstrate that the translation accuracy can be improved by allowing a smooth transaction between two RNN layers and using the state-of-the-art feature extractor. The experimental results on our new challenging dataset show that our approach outperforms recent methods by a fair margin. Furthermore, we combine the proposed translation module with the vision and planning system to let a robot perform various manipulation tasks. Finally, we demonstrate the effectiveness of our framework on a full-size humanoid robot WALK-MAN.

cs.RO↗

arXiv2017

3DTouch: Towards a Wearable 3D Input Device for 3D Applications

Three-dimensional (3D) applications have come to every corner of life. We present 3DTouch, a novel 3D wearable input device worn on the fingertip for interacting with 3D applications. 3DTouch is self-contained, and designed to universally work on various 3D platforms. The device employs touch input for the benefits of passive haptic feedback, and movement stability. Moreover, with touch interaction, 3DTouch is conceptually less fatiguing to use over many hours than 3D spatial input devices such as Kinect. Our approach relies on relative positioning technique using an optical laser sensor and a 9-DOF inertial measurement unit. We implemented a set of 3D interaction techniques including selection, translation, and rotation using 3DTouch. An evaluation also demonstrates the device's tracking accuracy of 1.10 mm and 2.33 degrees for subtle touch interaction in 3D space. With 3DTouch project, we would like to provide an input device that reduces the gap between 3D applications and users.

cs.HC↗

arXiv2017

Plug & Play Generative Networks: Conditional Iterative Generation of Images in Latent Space

Generating high-resolution, photo-realistic images has been a long-standing goal in machine learning. Recently, Nguyen et al. (2016) showed one interesting way to synthesize novel images by performing gradient ascent in the latent space of a generator network to maximize the activations of one or multiple neurons in a separate classifier network. In this paper we extend this method by introducing an additional prior on the latent code, improving both sample quality and sample diversity, leading to a state-of-the-art generative model that produces high quality images at higher resolutions (227x227) than previous generative models, and does so for all 1000 ImageNet categories. In addition, we provide a unified probabilistic interpretation of related activation maximization methods and call the general class of models "Plug and Play Generative Networks". PPGNs are composed of 1) a generator network G that is capable of drawing a wide range of image types and 2) a replaceable "condition" network C that tells the generator what to draw. We demonstrate the generation of images conditioned on a class (when C is an ImageNet or MIT Places classification network) and also conditioned on a caption (when C is an image captioning network). Our method also improves the state of the art of Multifaceted Feature Visualization, which generates the set of synthetic inputs that activate a neuron in order to better understand how deep neural networks operate. Finally, we show that our model performs reasonably well at the task of image inpainting. While image models are used in this paper, the approach is modality-agnostic and can be applied to many types of data.

cs.CV↗

arXiv2016

"420 Friendly": Revealing Marijuana Use via Craigslist Rental Ads

Recent studies have shown that information mined from Craigslist can be used for informing public health policy or monitoring risk behavior. This paper presents a text-mining method for conducting public health surveillance of marijuana use concerns in the U.S. using online classified ads in Craigslist. We collected more than 200 thousands of rental ads in the housing categories in Craigslist and devised text-mining methods for efficiently and accurately extract rental ads associated with concerns about the uses of marijuana in different states across the U.S. We linked the extracted ads to their geographic locations and computed summary statistics of the ads having marijuana use concerns. Our data is then compared with the State Marijuana Laws Map published by the U.S. government and marijuana related keywords search in Google to verify our collected data with respect to the demographics of marijuana use concerns. Our data not only indicates strong correlations between Craigslist ads, Google search and the State Marijuana Laws Map in states where marijuana uses are legal, but also reveals some hidden world of marijuana use concerns in other states where marijuana use is illegal. Our approach can be utilized as a marijuana surveillance tool for policy makers to develop public health policy and regulations.

cs.CY↗

arXiv2016

Evaluating Marijuana-Related Tweets On Twitter

This paper studies marijuana-related tweets in social network Twitter. We collected more than 300,000 marijuana related tweets during November 2016 in our study. Our text-mining based algorithms and data analysis unveil some interesting patterns including: (i) users' attitudes (e.g., positive or negative) can be characterized by the existence of outer links in a tweet; (ii) 67% users use their mobile phones to post their messages while many users publish their messages using third-party automatic posting services; and (3) the number of tweets during weekends is much higher than during weekdays. Our data also showed the impact of the political events such as the U.S. presidential election or state marijuana legalization votes on the marijuana-related tweeting frequencies.

cs.SI↗

arXiv2016

Joint Network Coding and Machine Learning for Error-prone Wireless Broadcast

Reliable broadcasting data to multiple receivers over lossy wireless channels is challenging due to the heterogeneity of the wireless link conditions. Automatic Repeat-reQuest (ARQ) based retransmission schemes are bandwidth inefficient due to data duplication at receivers. Network coding (NC) has been shown to be a promising technique for improving network bandwidth efficiency by combining multiple lost data packets for retransmission. However, it is challenging to accurately determine which lost packets should be combined together due to disrupted feedback channels. This paper proposes an adaptive data encoding scheme at the transmitter by joining network coding and machine learning (NCML) for retransmission of lost packets. Our proposed NCML extracts the important features from historical feedback signals received by the transmitter to train a classifier. The constructed classifier is then used to predict states of transmitted data packets at different receivers based on their corrupted feedback signals for effective data mixing. We have conducted extensive simulations to collaborate the efficiency of our proposed approach. The simulation results show that our machine learning algorithm can be trained efficiently and accurately. The simulation results show that on average the proposed NCML can correctly classify 90% of the states of transmitted data packets at different receivers. It achieves significant bandwidth gain compared with the ARQ and NC based schemes in different transmission terrains, power levels, and the distances between the transmitter and receivers.

cs.NI↗

arXiv2016

Synthesizing the preferred inputs for neurons in neural networks via deep generator networks

Deep neural networks (DNNs) have demonstrated state-of-the-art results on many pattern recognition tasks, especially vision classification problems. Understanding the inner workings of such computational brains is both fascinating basic science that is interesting in its own right - similar to why we study the human brain - and will enable researchers to further improve DNNs. One path to understanding how a neural network functions internally is to study what each of its neurons has learned to detect. One such method is called activation maximization (AM), which synthesizes an input (e.g. an image) that highly activates a neuron. Here we dramatically improve the qualitative state of the art of activation maximization by harnessing a powerful, learned prior: a deep generator network (DGN). The algorithm (1) generates qualitatively state-of-the-art synthetic images that look almost real, (2) reveals the features learned by each neuron in an interpretable way, (3) generalizes well to new datasets and somewhat well to different network architectures without requiring the prior to be relearned, and (4) can be considered as a high-quality generative method (in this case, by generating novel, creative, interesting, recognizable images).

cs.NE↗

arXiv2016

Multifaceted Feature Visualization: Uncovering the Different Types of Features Learned By Each Neuron in Deep Neural Networks

We can better understand deep neural networks by identifying which features each of their neurons have learned to detect. To do so, researchers have created Deep Visualization techniques including activation maximization, which synthetically generates inputs (e.g. images) that maximally activate each neuron. A limitation of current techniques is that they assume each neuron detects only one type of feature, but we know that neurons can be multifaceted, in that they fire in response to many different types of features: for example, a grocery store class neuron must activate either for rows of produce or for a storefront. Previous activation maximization techniques constructed images without regard for the multiple different facets of a neuron, creating inappropriate mixes of colors, parts of objects, scales, orientations, etc. Here, we introduce an algorithm that explicitly uncovers the multiple facets of each neuron by producing a synthetic visualization of each of the types of images that activate a neuron. We also introduce regularization methods that produce state-of-the-art results in terms of the interpretability of images obtained by activation maximization. By separately synthesizing each type of image a neuron fires in response to, the visualizations have more appropriate colors and coherent global structure. Multifaceted feature visualization thus provides a clearer and more comprehensive description of the role of each neuron.

cs.NE↗

arXiv2015

3DTouch: A wearable 3D input device with an optical sensor and a 9-DOF inertial measurement unit

We present 3DTouch, a novel 3D wearable input device worn on the fingertip for 3D manipulation tasks. 3DTouch is designed to fill the missing gap of a 3D input device that is self-contained, mobile, and universally working across various 3D platforms. This paper presents a low-cost solution to designing and implementing such a device. Our approach relies on relative positioning technique using an optical laser sensor and a 9-DOF inertial measurement unit. 3DTouch is self-contained, and designed to universally work on various 3D platforms. The device employs touch input for the benefits of passive haptic feedback, and movement stability. On the other hand, with touch interaction, 3DTouch is conceptually less fatiguing to use over many hours than 3D spatial input devices. We propose a set of 3D interaction techniques including selection, translation, and rotation using 3DTouch. An evaluation also demonstrates the device's tracking accuracy of 1.10 mm and 2.33 degrees for subtle touch interaction in 3D space. Modular solutions like 3DTouch opens up a whole new design space for interaction techniques to further develop on.

cs.HC↗

arXiv2015

Understanding Neural Networks Through Deep Visualization

Recent years have produced great advances in training large, deep neural networks (DNNs), including notable successes in training convolutional neural networks (convnets) to recognize natural images. However, our understanding of how these models work, especially what computations they perform at intermediate layers, has lagged behind. Progress in the field will be further accelerated by the development of better tools for visualizing and interpreting neural nets. We introduce two such tools here. The first is a tool that visualizes the activations produced on each layer of a trained convnet as it processes an image or video (e.g. a live webcam stream). We have found that looking at live activations that change in response to user input helps build valuable intuitions about how convnets work. The second tool enables visualizing features at each layer of a DNN via regularized optimization in image space. Because previous versions of this idea produced less recognizable images, here we introduce several new regularization methods that combine to produce qualitatively clearer, more interpretable visualizations. Both tools are open source and work on a pre-trained convnet with minimal setup.

cs.CV↗

arXiv2015

Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images

Deep neural networks (DNNs) have recently been achieving state-of-the-art performance on a variety of pattern-recognition tasks, most notably visual classification problems. Given that DNNs are now able to classify objects in images with near-human-level performance, questions naturally arise as to what differences remain between computer and human vision. A recent study revealed that changing an image (e.g. of a lion) in a way imperceptible to humans can cause a DNN to label the image as something else entirely (e.g. mislabeling a lion a library). Here we show a related result: it is easy to produce images that are completely unrecognizable to humans, but that state-of-the-art DNNs believe to be recognizable objects with 99.99% confidence (e.g. labeling with certainty that white noise static is a lion). Specifically, we take convolutional neural networks trained to perform well on either the ImageNet or MNIST datasets and then find images with evolutionary algorithms or gradient ascent that DNNs label with high confidence as belonging to each dataset class. It is possible to produce images totally unrecognizable to human eyes that DNNs believe with near certainty are familiar objects, which we call "fooling images" (more generally, fooling examples). Our results shed light on interesting differences between human vision and current DNNs, and raise questions about the generality of DNN computer vision.

cs.CV↗