arXiv subjects

Sheng Guo

Publications and source records attributed to Sheng Guo.

62 records · Page 4Linked to original sources

arXiv2017

Knowledge Guided Disambiguation for Large-Scale Scene Classification with Multi-Resolution CNNs

Convolutional Neural Networks (CNNs) have made remarkable progress on scene recognition, partially due to these recent large-scale scene datasets, such as the Places and Places2. Scene categories are often defined by multi-level information, including local objects, global layout, and background environment, thus leading to large intra-class variations. In addition, with the increasing number of scene categories, label ambiguity has become another crucial issue in large-scale classification. This paper focuses on large-scale scene recognition and makes two major contributions to tackle these issues. First, we propose a multi-resolution CNN architecture that captures visual content and structure at multiple levels. The multi-resolution CNNs are composed of coarse resolution CNNs and fine resolution CNNs, which are complementary to each other. Second, we design two knowledge guided disambiguation techniques to deal with the problem of label ambiguity. (i) We exploit the knowledge from the confusion matrix computed on validation data to merge ambiguous classes into a super category. (ii) We utilize the knowledge of extra networks to produce a soft label for each image. Then the super categories or soft labels are employed to guide CNN training on the Places2. We conduct extensive experiments on three large-scale image datasets (ImageNet, Places, and Places2), demonstrating the effectiveness of our approach. Furthermore, our method takes part in two major scene recognition challenges, and achieves the second place at the Places2 challenge in ILSVRC 2015, and the first place at the LSUN challenge in CVPR 2016. Finally, we directly test the learned representations on other scene benchmarks, and obtain the new state-of-the-art results on the MIT Indoor67 (86.7\%) and SUN397 (72.0\%). We release the code and models at~\url{https://github.com/wanglimin/MRCNN-Scene-Recognition}.

cs.CV↗

arXiv2016

Locally-Supervised Deep Hybrid Model for Scene Recognition

Convolutional neural networks (CNN) have recently achieved remarkable successes in various image classification and understanding tasks. The deep features obtained at the top fully-connected layer of the CNN (FC-features) exhibit rich global semantic information and are extremely effective in image classification. On the other hand, the convolutional features in the middle layers of the CNN also contain meaningful local information, but are not fully explored for image representation. In this paper, we propose a novel Locally-Supervised Deep Hybrid Model (LS-DHM) that effectively enhances and explores the convolutional features for scene recognition. Firstly, we notice that the convolutional features capture local objects and fine structures of scene images, which yield important cues for discriminating ambiguous scenes, whereas these features are significantly eliminated in the highly-compressed FC representation. Secondly, we propose a new Local Convolutional Supervision (LCS) layer to enhance the local structure of the image by directly propagating the label information to the convolutional layers. Thirdly, we propose an efficient Fisher Convolutional Vector (FCV) that successfully rescues the orderless mid-level semantic information (e.g. objects and textures) of scene image. The FCV encodes the large-sized convolutional maps into a fixed-length mid-level representation, and is demonstrated to be strongly complementary to the high-level FC-features. Finally, both the FCV and FC-features are collaboratively employed in the LSDHM representation, which achieves outstanding performance in our experiments. It obtains 83.75% and 67.56% accuracies respectively on the heavily benchmarked MIT Indoor67 and SUN397 datasets, advancing the stat-of-the-art substantially.

cs.CV↗

arXiv2016

Combining internally contracted states and matrix product states to perform multireference perturbation theory

We present two efficient and intruder-free methods for treating dynamic correlation on top of general multi-configuration reference wave functions---including such as obtained by the density matrix renormalization group (DMRG) with large active spaces. The new methods are the second order variant of the recently proposed multi-reference linearized coupled cluster method (MRLCC) [S. Sharma, A. Alavi, J. Chem. Phys. 143, 102815 (2015)], and of N-electron valence perturbation theory (NEVPT2), with expected accuracies similar to MRCI+Q and (at least) CASPT2, respectively. Great efficiency gains are realized by representing the first-order wave function with a combination of internal contraction (IC) and matrix product state perturbation theory (MPSPT). With this combination, only third order reduced density matrices (RDMs) are required. Thus, we obviate the need for calculating (or estimating) RDMs of fourth or higher order; these had so far posed a severe bottleneck for dynamic correlation treatments involving the large active spaces accessible to DMRG. Using several benchmark systems, including first and second row containing small molecules, Cr$_2$, pentacene and oxo-Mn(Salen), we shown that active spaces containing at least 30 orbitals can be treated using this method. On a single node, MRLCC2 and NEVPT2 calculations can be performed with over 550 and 1100 virtual orbitals, respectively. We also critically examine the errors incurred due to the three sources of errors introduced in the present implementation - calculating second order instead of third order energy corrections, use of internal contraction and approximations made in the reference wavefunction due to DMRG.

physics.chem-ph↗

arXiv2016

N-electron valence state perturbation theory based on a density matrix renormalization group reference function, with applications to the chromium dimer and poly-p-phenylene vinylene oligomer

The strongly-contracted variant of second order N -electron valence state perturbation theory (NEVPT2) is an efficient perturbative method to treat dynamic correlation without the problems of intruder states or level shifts, while the density matrix renormalization group (DMRG) provides the capability to tackle static correlation in large active spaces. We present a combination of the DMRG and strongly-contracted NEVPT2 (DMRG-SC-NEVPT2) that uses an efficient algorithm to compute high order reduced density matrices from DMRG wave functions. The capabilities of DMRG-SC-NEVPT2 are demonstrated on calculations of the chromium dimer potential energy curve at the basis set limit, and the excitation energies of poly-p-phenylene vinylene trimer (PPV(n=3)).

physics.chem-ph↗

arXiv2015

Better Exploiting OS-CNNs for Better Event Recognition in Images

Event recognition from still images is one of the most important problems for image understanding. However, compared with object recognition and scene recognition, event recognition has received much less research attention in computer vision community. This paper addresses the problem of cultural event recognition in still images and focuses on applying deep learning methods on this problem. In particular, we utilize the successful architecture of Object-Scene Convolutional Neural Networks (OS-CNNs) to perform event recognition. OS-CNNs are composed of object nets and scene nets, which transfer the learned representations from the pre-trained models on large-scale object and scene recognition datasets, respectively. We propose four types of scenarios to explore OS-CNNs for event recognition by treating them as either "end-to-end event predictors" or "generic feature extractors". Our experimental results demonstrate that the global and local representations of OS-CNNs are complementary to each other. Finally, based on our investigation of OS-CNNs, we come up with a solution for the cultural event recognition track at the ICCV ChaLearn Looking at People (LAP) challenge 2015. Our team secures the third place at this challenge and our result is very close to the best performance.

cs.CV↗

arXiv2015

Places205-VGGNet Models for Scene Recognition

VGGNets have turned out to be effective for object recognition in still images. However, it is unable to yield good performance by directly adapting the VGGNet models trained on the ImageNet dataset for scene recognition. This report describes our implementation of training the VGGNets on the large-scale Places205 dataset. Specifically, we train three VGGNet models, namely VGGNet-11, VGGNet-13, and VGGNet-16, by using a Multi-GPU extension of Caffe toolbox with high computational efficiency. We verify the performance of trained Places205-VGGNet models on three datasets: MIT67, SUN397, and Places205. Our trained models achieve the state-of-the-art performance on these datasets and are made public available.

cs.CV↗

arXiv2015

Local Color Contrastive Descriptor for Image Classification

Image representation and classification are two fundamental tasks towards multimedia content retrieval and understanding. The idea that shape and texture information (e.g. edge or orientation) are the key features for visual representation is ingrained and dominated in current multimedia and computer vision communities. A number of low-level features have been proposed by computing local gradients (e.g. SIFT, LBP and HOG), and have achieved great successes on numerous multimedia applications. In this paper, we present a simple yet efficient local descriptor for image classification, referred as Local Color Contrastive Descriptor (LCCD), by leveraging the neural mechanisms of color contrast. The idea originates from the observation in neural science that color and shape information are linked inextricably in visual cortical processing. The color contrast yields key information for visual color perception and provides strong linkage between color and shape. We propose a novel contrastive mechanism to compute the color contrast in both spatial location and multiple channels. The color contrast is computed by measuring \emph{f}-divergence between the color distributions of two regions. Our descriptor enriches local image representation with both color and contrast information. We verified experimentally that it can compensate strongly for the shape based descriptor (e.g. SIFT), while keeping computationally simple. Extensive experimental results on image classification show that our descriptor improves the performance of SIFT substantially by combinations, and achieves the state-of-the-art performance on three challenging benchmark datasets. It improves recent Deep Learning model (DeCAF) [1] largely from the accuracy of 40.94% to 49.68% in the large scale SUN397 database. Codes for the LCCD will be available.

cs.CV↗

arXiv2009

Large-scale simulation of RNA macroevolution by an energy-dependent fitness model

Simulated nucleotide sequences are widely used in theoretical and empirical molecular evolution studies. Conventional simulators generally use fixed parameter time-homogeneous Markov model for sequence evolution. In this work, we use the folding free energy of the secondary structure of an RNA as a proxy for its phenotypic fitness, and simulate RNA macroevolution by a mutation-selection population genetics model. Because the two-step process is conditioned on an RNA and its mutant ensemble, we no longer have a global substitution matrix, nor do we explicitly assume any for this inhomogeneous stochastic process. After introducing the base model of RNA evolution, we outline the heuristic implementation algorithm and several model improvements. We then discuss the calibration of the model parameters and demonstrate that in phylogeny reconstruction with both the parsimony method and the likelihood method, the sequences generated by our simulator, rnasim, have greater statistical complexity than those by two standard simulators, ROSE and Seq-Gen, and are close to empirical sequences.

q-bio.PE↗