arXiv subjects

Qing Liu

Publications and source records attributed to Qing Liu.

At least 199 records · Page 11Linked to original sources

arXiv2020

Dust Attenuation Curve for Local Subgalactic Star-forming Regions

We compile a sample of about 157,000 spaxels from the Mapping Nearby Galaxies at the Apache Point Observatory survey to derive the average dust attenuation curve for subgalactic star-forming regions of local star-forming galaxies (SFGs) in the optical wavelength, following the method of \cite{Calzetti1994}. We obtain a $D_n(4000)$-independent average attenuation curve for spaxels with $1.1\leq D_n(4000)<1.3$, which is similar to the one derived from either local starbursts or normal SFGs. We examine whether and how the shape of the average attenuation curve changes with several local and global physical properties. For spaxels with $1.2\leq D_n(4000)<1.3$, we find no dependence on either local or global physical properties for the shape of the average attenuation curve. However, for spaxels with younger stellar population ($1.1\leq D_n(4000)<1.2$), shallower average attenuation curves are found for star-forming regions with smaller stellar mass surface density, smaller star formation rate surface density, or those residing in the outer region of galaxies. These results emphasize the risk of using one single attenuation curve to correct the dust reddening for all types of star-forming regions, especially for those with fairly young stellar population.

astro-ph.GA↗

arXiv2020

U-rank: Utility-oriented Learning to Rank with Implicit Feedback

Learning to rank with implicit feedback is one of the most important tasks in many real-world information systems where the objective is some specific utility, e.g., clicks and revenue. However, we point out that existing methods based on probabilistic ranking principle do not necessarily achieve the highest utility. To this end, we propose a novel ranking framework called U-rank that directly optimizes the expected utility of the ranking list. With a position-aware deep click-through rate prediction model, we address the attention bias considering both query-level and item-level features. Due to the item-specific attention bias modeling, the optimization for expected utility corresponds to a maximum weight matching on the item-position bipartite graph. We base the optimization of this objective in an efficient Lambdaloss framework, which is supported by both theoretical and empirical analysis. We conduct extensive experiments for both web search and recommender systems over three benchmark datasets and two proprietary datasets, where the performance gain of U-rank over state-of-the-arts is demonstrated. Moreover, our proposed U-rank has been deployed on a large-scale commercial recommender and a large improvement over the production baseline has been observed in an online A/B testing.

cs.IR↗

arXiv2020

Reduction type of smooth quartics

Let $C/K$ be a smooth plane quartic over a discrete valuation field. We characterize the type of reduction (i.e. smooth plane quartic, hyperelliptic genus 3 curve or bad) over $K$ in terms of the existence of a special plane quartic model and, over $\bar{K}$, in terms of the valuations of certain algebraic invariants of $C$ when the characteristic of the residue field is not $2,\,3,\,5$ or $7$. On the way, we gather several results of general interest on geometric invariant theory over an arbitrary ring $R$ in the spirit of (Seshadri 1977). For instance when $R$ is a discrete valuation ring, we show the existence of a homogeneous system of parameters over $R$. We exhibit explicit ones for ternary quartic forms under the action of $\textrm{SL}_{3,R}$ depending only on the characteristic $p$ of the residue field. We illustrate our results with the case of Picard curves for which we give simple criteria for the type of reduction.

math.NT↗

arXiv2020

Compositional Convolutional Neural Networks: A Robust and Interpretable Model for Object Recognition under Occlusion

Computer vision systems in real-world applications need to be robust to partial occlusion while also being explainable. In this work, we show that black-box deep convolutional neural networks (DCNNs) have only limited robustness to partial occlusion. We overcome these limitations by unifying DCNNs with part-based models into Compositional Convolutional Neural Networks (CompositionalNets) - an interpretable deep architecture with innate robustness to partial occlusion. Specifically, we propose to replace the fully connected classification head of DCNNs with a differentiable compositional model that can be trained end-to-end. The structure of the compositional model enables CompositionalNets to decompose images into objects and context, as well as to further decompose object representations in terms of individual parts and the objects' pose. The generative nature of our compositional model enables it to localize occluders and to recognize objects based on their non-occluded parts. We conduct extensive experiments in terms of image classification and object detection on images of artificially occluded objects from the PASCAL3D+ and ImageNet dataset, and real images of partially occluded vehicles from the MS-COCO dataset. Our experiments show that CompositionalNets made from several popular DCNN backbones (VGG-16, ResNet50, ResNext) improve by a large margin over their non-compositional counterparts at classifying and detecting partially occluded objects. Furthermore, they can localize occluders accurately despite being trained with class-level supervision only. Finally, we demonstrate that CompositionalNets provide human interpretable predictions as their individual components can be understood as detecting parts and estimating an objects' viewpoint.

cs.CV↗

arXiv2020

Equivalence of solutions of eikonal equation in metric spaces

In this paper we prove the equivalence between some known notions of solutions to the eikonal equation and more general analogs of the Hamilton-Jacobi equations in complete and rectifiably connected metric spaces. The notions considered are that of curve-based viscosity solutions, slope-based viscosity solutions, and Monge solutions. By using the induced intrinsic (path) metric, we reduce the metric space to a length space and show the equivalence of these solutions to the associated Dirichlet boundary problem. Without utilizing the boundary data, we also localize our argument and directly prove the equivalence for the definitions of solutions. Regularity of solutions related to the Euclidean semi-concavity is discussed as well.

math.AP↗

arXiv2020

Maps on the Morse boundary

For a proper geodesic metric space $X$, the Morse boundary $\partial_*X$ focuses on the hyperbolic-like directions in the space $X$. It is a quasi-isometry invariant. That is, a quasi-isometry between two hyperbolic spaces induces a homeomorphism on their boundaries. In this paper, we investigate additional structures on the Morse boundary $\partial_*X$ which determine $X$ up to a quasi-isometry. We prove that, for $X$ and $Y$ proper, cocompact spaces, a homeomorphism $f$ between their Morse boundaries is induced by a quasi-isometry if and only if $f$ and $f^{-1}$ are bihölder, or quasi-symmetric, or strongly quasi-conformal.

math.GT↗

arXiv2020

Incremental Meta-Learning via Indirect Discriminant Alignment

Majority of the modern meta-learning methods for few-shot classification tasks operate in two phases: a meta-training phase where the meta-learner learns a generic representation by solving multiple few-shot tasks sampled from a large dataset and a testing phase, where the meta-learner leverages its learnt internal representation for a specific few-shot task involving classes which were not seen during the meta-training phase. To the best of our knowledge, all such meta-learning methods use a single base dataset for meta-training to sample tasks from and do not adapt the algorithm after meta-training. This strategy may not scale to real-world use-cases where the meta-learner does not potentially have access to the full meta-training dataset from the very beginning and we need to update the meta-learner in an incremental fashion when additional training data becomes available. Through our experimental setup, we develop a notion of incremental learning during the meta-training phase of meta-learning and propose a method which can be used with multiple existing metric-based meta-learning algorithms. Experimental results on benchmark dataset show that our approach performs favorably at test time as compared to training a model with the full meta-training set and incurs negligible amount of catastrophic forgetting

cs.LG↗

arXiv2020

The Dragonfly Wide Field Survey. I. Telescope, Survey Design and Data Characterization

We present a description of the Dragonfly Wide Field Survey (DWFS), a deep photometric survey of a wide area of sky. The DWFS covers 330 $\mathrm{deg}^2$ in the equatorial GAMA fields and the Stripe 82 fields in the SDSS $g$ and $r$ bands. It is carried out with the 48-lens Dragonfly Telephoto Array, a telescope that is optimized for the detection of low surface brightness emission. The main goal of the survey is to study the dwarf galaxy population beyond the Local Group. In this paper, we describe the survey design and show early results. We reach $1σ$ depths of $μ_g\approx 31$ mag arcsec$^{-2}$ on arcminute scales and show that Milky Way satellites such as Sextans, Bootes, and Ursa Major should be detectable out to $D\gtrsim 10$ Mpc. We also provide an overview of the elements and operation of the 48-lens Dragonfly telescope and a detailed description of its data reduction pipeline. The pipeline is fully automated, with individual frames subjected to a rigorous series of quality tests. The sky subtraction is performed in two stages, ensuring that emission features with spatial scales up to $\sim 0.^{\circ}9 \times 0.^{\circ}6$ are preserved. The DWFS provides unparalleled sensitivity to low surface brightness features on arcminute scales.

astro-ph.GA↗

arXiv2020

Compositional Convolutional Neural Networks: A Deep Architecture with Innate Robustness to Partial Occlusion

Recent findings show that deep convolutional neural networks (DCNNs) do not generalize well under partial occlusion. Inspired by the success of compositional models at classifying partially occluded objects, we propose to integrate compositional models and DCNNs into a unified deep model with innate robustness to partial occlusion. We term this architecture Compositional Convolutional Neural Network. In particular, we propose to replace the fully connected classification head of a DCNN with a differentiable compositional model. The generative nature of the compositional model enables it to localize occluders and subsequently focus on the non-occluded parts of the object. We conduct classification experiments on artificially occluded images as well as real images of partially occluded objects from the MS-COCO dataset. The results show that DCNNs do not classify occluded objects robustly, even when trained with data that is strongly augmented with partial occlusions. Our proposed model outperforms standard DCNNs by a large margin at classifying partially occluded objects, even when it has not been exposed to occluded objects during training. Additional experiments demonstrate that CompositionalNets can also localize the occluders accurately, despite being trained with class labels only. The code used in this work is publicly available.

cs.CV↗

arXiv2020

Adversarial Attack on DL-based Massive MIMO CSI Feedback

With the increasing application of deep learning (DL) algorithms in wireless communications, the physical layer faces new challenges caused by adversarial attack. Such attack has significantly affected the neural network in computer vision. We chose DL-based analog channel state information (CSI) to show the effect of adversarial attack on DL-based communication system. We present a practical method to craft white-box adversarial attack on DL-based CSI feedback process. Our simulation results showed the destructive effect adversarial attack caused on DL-based CSI feedback by analyzing the performance of normalized mean square error. We also launched a jamming attack for comparison and found that the jamming attack could be prevented with certain precautions. As DL algorithm becomes the trend in developing wireless communication, this work raises concerns regarding the security in the use of DL-based algorithms.

cs.IT↗

arXiv2020

Combining Compositional Models and Deep Networks For Robust Object Classification under Occlusion

Deep convolutional neural networks (DCNNs) are powerful models that yield impressive results at object classification. However, recent work has shown that they do not generalize well to partially occluded objects and to mask attacks. In contrast to DCNNs, compositional models are robust to partial occlusion, however, they are not as discriminative as deep models. In this work, we combine DCNNs and compositional object models to retain the best of both approaches: a discriminative model that is robust to partial occlusion and mask attacks. Our model is learned in two steps. First, a standard DCNN is trained for image classification. Subsequently, we cluster the DCNN features into dictionaries. We show that the dictionary components resemble object part detectors and learn the spatial distribution of parts for each object class. We propose mixtures of compositional models to account for large changes in the spatial activation patterns (e.g. due to changes in the 3D pose of an object). At runtime, an image is first classified by the DCNN in a feedforward manner. The prediction uncertainty is used to detect partially occluded objects, which in turn are classified by the compositional model. Our experimental results demonstrate that combining compositional models and DCNNs resolves a fundamental problem of current deep learning approaches to computer vision: The combined model recognizes occluded objects, even when it has not been exposed to occluded objects during training, while at the same time maintaining high discriminative performance for non-occluded objects.

cs.CV↗

arXiv2019

Comparison-Based Convolutional Neural Networks for Cervical Cell/Clumps Detection in the Limited Data Scenario

Automated detection of cervical cancer cells or cell clumps has the potential to significantly reduce error rate and increase productivity in cervical cancer screening. However, most traditional methods rely on the success of accurate cell segmentation and discriminative hand-crafted features extraction. Recently there are emerging deep learning-based methods which train convolutional neural networks (CNN) to classify image patches, but they are computationally expensive. In this paper we propose an efficient CNN-based object detection methods for cervical cancer cells/clumps detection. Specifically, we utilize the state-of-the-art two-stage object detection method, the Faster-RCNN with Feature Pyramid Network (FPN) as the baseline and propose a novel comparison detector to deal with the limited data problem. The key idea is that classify the proposals by comparing with the reference samples of each category in object detection. In addition, we propose to learn the reference samples of the background from data instead of manually choosing them by some heuristic rules. Experimental results show that the proposed Comparison Detector yields significant improvement on the small dataset, achieving a mean Average Precision (mAP) of 26.3% and an Average Recall (AR) of 35.7%, both improving about 20 points compared to the baseline. Moreover, Comparison Detector improved AR by 4.6 points and achieved marginally better performance in terms of mAP compared with baseline model when training on the medium dataset. Our method is promising for the development of automation-assisted cervical cancer screening systems. Code is available at https://github.com/kuku-sichuan/ComparisonDetector.

cs.CV↗

arXiv2019

Localizing Occluders with Compositional Convolutional Networks

Compositional convolutional networks are generative compositional models of neural network features, that achieve state of the art results when classifying partially occluded objects, even when they have not been exposed to occluded objects during training. In this work, we study the performance of CompositionalNets at localizing occluders in images. We show that the original model is not able to localize occluders well. We propose to overcome this limitation by modeling the feature activations as a mixture of von-Mises-Fisher distributions, which also allows for an end-to-end training of CompositionalNets. Our experimental results demonstrate that the proposed extensions increase the model's performance at localizing occluders as well as at classifying partially occluded objects.

cs.CV↗

arXiv2019

A Deep Gradient Boosting Network for Optic Disc and Cup Segmentation

Segmentation of optic disc (OD) and optic cup (OC) is critical in automated fundus image analysis system. Existing state-of-the-arts focus on designing deep neural networks with one or multiple dense prediction branches. Such kind of designs ignore connections among prediction branches and their learning capacity is limited. To build connections among prediction branches, this paper introduces gradient boosting framework to deep classification model and proposes a gradient boosting network called BoostNet. Specifically, deformable side-output unit and aggregation unit with deep supervisions are proposed to learn base functions and expansion coefficients in gradient boosting framework. By stacking aggregation units in a deep-to-shallow manner, models' performances are gradually boosted along deep to shallow stages. BoostNet achieves superior results to existing deep OD and OC segmentation networks on the public dataset ORIGA.

cs.CV↗

arXiv2019

Semantic-Aware Knowledge Preservation for Zero-Shot Sketch-Based Image Retrieval

Sketch-based image retrieval (SBIR) is widely recognized as an important vision problem which implies a wide range of real-world applications. Recently, research interests arise in solving this problem under the more realistic and challenging setting of zero-shot learning. In this paper, we investigate this problem from the viewpoint of domain adaptation which we show is critical in improving feature embedding in the zero-shot scenario. Based on a framework which starts with a pre-trained model on ImageNet and fine-tunes it on the training set of SBIR benchmark, we advocate the importance of preserving previously acquired knowledge, e.g., the rich discriminative features learned from ImageNet, to improve the model's transfer ability. For this purpose, we design an approach named Semantic-Aware Knowledge prEservation (SAKE), which fine-tunes the pre-trained model in an economical way and leverages semantic information, e.g., inter-class relationship, to achieve the goal of knowledge preservation. Zero-shot experiments on two extended SBIR datasets, TU-Berlin and Sketchy, verify the superior performance of our approach. Extensive diagnostic experiments validate that knowledge preserved benefits SBIR in zero-shot settings, as a large fraction of the performance gain is from the more properly structured feature embedding for photo images. Code is available at: https://github.com/qliu24/SAKE.

cs.CV↗

arXiv2019

Semantic Part Detection via Matching: Learning to Generalize to Novel Viewpoints from Limited Training Data

Detecting semantic parts of an object is a challenging task in computer vision, particularly because it is hard to construct large annotated datasets due to the difficulty of annotating semantic parts. In this paper we present an approach which learns from a small training dataset of annotated semantic parts, where the object is seen from a limited range of viewpoints, but generalizes to detect semantic parts from a much larger range of viewpoints. Our approach is based on a matching algorithm for finding accurate spatial correspondence between two images, which enables semantic parts annotated on one image to be transplanted to another. In particular, this enables images in the training dataset to be matched to a virtual 3D model of the object (for simplicity, we assume that the object viewpoint can be estimated by standard techniques). Then a clustering algorithm is used to annotate the semantic parts of the 3D virtual model. This virtual 3D model can be used to synthesize annotated images from a large range of viewpoint. These can be matched to images in the test set, using the same matching algorithm, to detect semantic parts in novel viewpoints of the object. Our algorithm is very simple, intuitive, and contains very few parameters. We evaluate our approach in the car subclass of the VehicleSemanticPart dataset. We show it outperforms standard deep network approaches and, in particular, performs much better on novel viewpoints. For facilitating the future research, code is available: https://github.com/ytongbai/SemanticPartDetection

cs.CV↗

arXiv2019

Dual-attention Focused Module for Weakly Supervised Object Localization

The research on recognizing the most discriminative regions provides referential information for weakly supervised object localization with only image-level annotations. However, the most discriminative regions usually conceal the other parts of the object, thereby impeding entire object recognition and localization. To tackle this problem, the Dual-attention Focused Module (DFM) is proposed to enhance object localization performance. Specifically, we present a dual attention module for information fusion, consisting of a position branch and a channel one. In each branch, the input feature map is deduced into an enhancement map and a mask map, thereby highlighting the most discriminative parts or hiding them. For the position mask map, we introduce a focused matrix to enhance it, which utilizes the principle that the pixels of an object are continuous. Between these two branches, the enhancement map is integrated with the mask map, aiming at partially compensating the lost information and diversifies the features. With the dual-attention module and focused matrix, the entire object region could be precisely recognized with implicit information. We demonstrate outperforming results of DFM in experiments. In particular, DFM achieves state-of-the-art performance in localization accuracy in ILSVRC 2016 and CUB-200-2011.

cs.CV↗

arXiv2019

Horizontal convex envelope in the Heisenberg group and applications to sub-elliptic equations

This paper introduces in a natural way a notion of horizontal convex envelopes of continuous functions in the Heisenberg group. We provide a convexification process to find the envelope in a constructive manner. We also apply the convexification process to show h-convexity of viscosity solutions to a class of fully nonlinear elliptic equations in the Heisenberg group satisfying a certain symmetry condition. Our examples show that in general one cannot expect h-convexity of solutions without the symmetry condition.

math.AP↗