arXiv subjects

Lai Jiang

Publications and source records attributed to Lai Jiang.

45 records · Page 3Linked to original sources

arXiv2023

Progressive Training of A Two-Stage Framework for Video Restoration

As a widely studied task, video restoration aims to enhance the quality of the videos with multiple potential degradations, such as noises, blurs and compression artifacts. Among video restorations, compressed video quality enhancement and video super-resolution are two of the main tacks with significant values in practical scenarios. Recently, recurrent neural networks and transformers attract increasing research interests in this field, due to their impressive capability in sequence-to-sequence modeling. However, the training of these models is not only costly but also relatively hard to converge, with gradient exploding and vanishing problems. To cope with these problems, we proposed a two-stage framework including a multi-frame recurrent network and a single-frame transformer. Besides, multiple training strategies, such as transfer learning and progressive training, are developed to shorten the training time and improve the model performance. Benefiting from the above technical contributions, our solution wins two champions and a runner-up in the NTIRE 2022 super-resolution and quality enhancement of compressed video challenges. Code is available at https://github.com/ryanxingql/winner-ntire22-vqe.

cs.CV↗

arXiv2022

Box dimension of generalized affine fractal interpolation functions

Let $f$ be a generalized affine fractal interpolation function with vertical scaling function $S$. In this paper, we study $\dim_B Γf$, the box dimension of the graph of $f$, under the assumption that $S$ is a Lipschtz function. By introducing vertical scaling matrices, we estimate the upper bound and the lower bound of oscillations of $f$. As a result, we obtain explicit formula of $\dim_B Γf$ under certain constraint conditions.

math.MG↗

arXiv2022

NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video: Dataset, Methods and Results

This paper reviews the NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video. In this challenge, we proposed the LDV 2.0 dataset, which includes the LDV dataset (240 videos) and 95 additional videos. This challenge includes three tracks. Track 1 aims at enhancing the videos compressed by HEVC at a fixed QP. Track 2 and Track 3 target both the super-resolution and quality enhancement of HEVC compressed video. They require x2 and x4 super-resolution, respectively. The three tracks totally attract more than 600 registrations. In the test phase, 8 teams, 8 teams and 12 teams submitted the final results to Tracks 1, 2 and 3, respectively. The proposed methods and solutions gauge the state-of-the-art of super-resolution and quality enhancement of compressed video. The proposed LDV 2.0 dataset is available at https://github.com/RenYang-home/LDV_dataset. The homepage of this challenge (including open-sourced codes) is at https://github.com/RenYang-home/NTIRE22_VEnh_SR.

cs.CV↗

arXiv2019

Attention Based Glaucoma Detection: A Large-scale Database and CNN Model

Recently, the attention mechanism has been successfully applied in convolutional neural networks (CNNs), significantly boosting the performance of many computer vision tasks. Unfortunately, few medical image recognition approaches incorporate the attention mechanism in the CNNs. In particular, there exists high redundancy in fundus images for glaucoma detection, such that the attention mechanism has potential in improving the performance of CNN-based glaucoma detection. This paper proposes an attention-based CNN for glaucoma detection (AG-CNN). Specifically, we first establish a large-scale attention based glaucoma (LAG) database, which includes 5,824 fundus images labeled with either positive glaucoma (2,392) or negative glaucoma (3,432). The attention maps of the ophthalmologists are also collected in LAG database through a simulated eye-tracking experiment. Then, a new structure of AG-CNN is designed, including an attention prediction subnet, a pathological area localization subnet and a glaucoma classification subnet. Different from other attention-based CNN methods, the features are also visualized as the localized pathological area, which can advance the performance of glaucoma detection. Finally, the experiment results show that the proposed AG-CNN approach significantly advances state-of-the-art glaucoma detection.

cs.CV↗

arXiv2019

Predicting Video Saliency with Object-to-Motion CNN and Two-layer Convolutional LSTM

Over the past few years, deep neural networks (DNNs) have exhibited great success in predicting the saliency of images. However, there are few works that apply DNNs to predict the saliency of generic videos. In this paper, we propose a novel DNN-based video saliency prediction method. Specifically, we establish a large-scale eye-tracking database of videos (LEDOV), which provides sufficient data to train the DNN models for predicting video saliency. Through the statistical analysis of our LEDOV database, we find that human attention is normally attracted by objects, particularly moving objects or the moving parts of objects. Accordingly, we propose an object-to-motion convolutional neural network (OM-CNN) to learn spatio-temporal features for predicting the intra-frame saliency via exploring the information of both objectness and object motion. We further find from our database that there exists a temporal correlation of human attention with a smooth saliency transition across video frames. Therefore, we develop a two-layer convolutional long short-term memory (2C-LSTM) network in our DNN-based method, using the extracted features of OM-CNN as the input. Consequently, the inter-frame saliency maps of videos can be generated, which consider the transition of attention across video frames. Finally, the experimental results show that our method advances the state-of-the-art in video saliency prediction.

cs.CV↗

arXiv2017

Fully Bayesian Classification with Heavy-tailed Priors for Selection in High-dimensional Features with Grouping Structure

Feature selection is demanded in many modern scientific research problems that use high-dimensional data. A typical example is to find the most useful genes that are related to a certain disease (eg, cancer) from high-dimensional gene expressions. The expressions of genes have grouping structures, for example, a group of co-regulated genes that have similar biological functions tend to have similar expressions. Many statistical methods have been proposed to take the grouping structure into consideration in feature selection, including group LASSO, supervised group LASSO, and regression on group representatives. In this paper, we propose a fully Bayesian Robit regression method with heavy-tailed (sparsity) priors (shortened by FBRHT) for selecting features with grouping structure. The main features of FBRHT include that it discards more aggressively unrelated features than LASSO, and it can make feature selection within groups automatically without a pre-specified grouping structure. In this paper, we use simulated and real datasets to demonstrate that the predictive power of the sparse feature subsets selected by FBRHT are comparable with other much larger feature subsets selected by LASSO, group LASSO, supervised group LASSO, penalized logistic regression and random forest, and that the succinct feature subsets selected by FBRHT have significantly better predictive power than the feature subsets of the same size taken from the top features selected by the aforementioned methods.

stat.ME↗

arXiv2016

Enhanced charge ordering transition in doped CaFeO3 through steric templating

We report density functional theory (DFT) investigation of $B$-site doped CaFeO$_3$, a prototypical charge-ordered perovskite. At 290 K, CaFeO$_3$ undergoes a metal-insulator transition and a charge disproportionation reaction 2Fe$^{4+}$$\rightarrow$Fe$^{5+}$+Fe$^{3+}$. We observe that when Zr dopants occupy a (001) layer, the band gap of the resulting solid solution increases to 0.93 eV due to a 2D Jahn-Teller type distortion, where FeO$_6$ cages on the $xy$ plane elongate along $x$ and $y$ alternatively between neighboring Fe sites. Furthermore, we show that the rock-salt ordering of the Fe$^{5+}$ and Fe$^{3+}$ cations can be enhanced when the $B$-site dopants are arranged in a (111) plane due to a collective steric effect that facilitates the size discrepancy between the Fe$^{5+}$O$_6$ and Fe$^{3+}$O$_6$ octahedra and therefore gives rise to a larger band gap. The enhanced charge disproportionation in these solid solutions is verified by rigorously calculating the oxidation states of the Fe cations with different octahedral cage sizes. We therefore predict that the corresponding transition temperature will increase due to the enhanced charge ordering and larger band gap. The compositional, structural and electrical relationships exploited in this paper can be extended to a variety of perovskites and non-perovskite oxides providing guidance in structurally manipulating electrical properties of functional materials.

cond-mat.mtrl-sci↗

arXiv2014

Coupling between octahedral rotations and local polar displacements in WO3/ReO3 superlattices

We model short-period superlattices of WO$_3$ and ReO$_3$ with first-principles calculations. In fully-relaxed superlattices, we observe that octahedral tilts about an axis in the planes of the superlattices do not propagate from one material, despite the presence of the corner-shared oxygen atoms. However, we find that octahedral rotation is enhanced within WO$_3$ layers in cases in which strain couples with native antiferroelectric displacements of tungsten within their octahedral cages. Resulting structures remain antiferroelectric with low net global polarization. Thermodynamic analysis reveals that superlattices with sufficiently thick ReO$_3$ layers, the absolute number being three or more layers and the Re fraction $\geq 50\%$, tend to be more stable than the separated material phases and also show enhanced octahedral rotations in the WO$_3$ layers.

cond-mat.mtrl-sci↗

arXiv2012

Rigorous definition of oxidation states of ions in solids

We present justification and rigorous procedure for electron partitioning among atoms in extended systems. The method is based on wavefunction topology and the modern theory of polarization, rather than charge density partitioning or wavefunction projection, and, as such, re-formulates the concept of oxidation state without assuming real-space charge transfer between atoms. This formulation provides rigorous electrostatics of finite extent solids, including films and nanowires.

cond-mat.mtrl-sci↗