Yan Wang

Yan Wang
sensetime · research

Ph.D

About

26
Publications
10,195
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
233
Citations
Additional affiliations
September 2014 - January 2017
Tsinghua University
Position
  • PhD Student
Education
September 2014 - July 2019
Tsinghua University
Field of study
  • Public Safety
September 2010 - July 2014
Tsinghua University
Field of study
  • Engineering Physics

Publications

Publications (26)
Preprint
Full-text available
In the past years, learned image compression (LIC) has achieved remarkable performance. The recent LIC methods outperform VVC in both PSNR and MS-SSIM. However, the low bit-rate reconstructions of LIC suffer from artifacts such as blurring, color drifting and texture missing. Moreover, those varied artifacts make image quality metrics correlate bad...
Article
Fire is one of the most frequent and common emergencies threatening public safety and social development. Recently, intelligent fire detection technologies represented by convolutional neural networks (CNNs) have been widely concerned by academia and industry, substantially improving detection accuracy. However, CNN-based fire detection systems are...
Preprint
JPEG is a popular image compression method widely used by individuals, data center, cloud storage and network filesystems. However, most recent progress on image compression mainly focuses on uncompressed images while ignoring trillions of already-existing JPEG images. To compress these JPEG images adequately and restore them back to JPEG format lo...
Preprint
Full-text available
Recently, learned image compression techniques have achieved remarkable performance, even surpassing the best manually designed lossy image coders. They are promising to be large-scale adopted. For the sake of practicality, a thorough investigation of the architecture design of learned image compression, regarding both compression performance and r...
Preprint
Full-text available
It has been witnessed that learned image compression has outperformed conventional image coding techniques and tends to be practical in industrial applications. One of the most critical issues that need to be considered is the non-deterministic calculation, which makes the probability prediction cross-platform inconsistent and frustrates successful...
Preprint
Full-text available
Learned image compression is making good progress in recent years. Peak signal-to-noise ratio (PSNR) and multi-scale structural similarity (MS-SSIM) are the two most popular evaluation metrics. As different metrics only reflect certain aspects of human perception, works in this field normally optimize two models using PSNR and MS-SSIM as loss funct...
Conference Paper
For learned image compression, the autoregressive context model is proved effective in improving the rate-distortion (RD) performance. Because it helps remove spatial redundancies among latent representations. However, the decoding process must be done in a strict scan order, which breaks the parallelization. We propose a parallelizable checkerboar...
Preprint
Full-text available
For learned image compression, the autoregressive context model is proved effective in improving the rate-distortion (RD) performance. Because it helps remove spatial redundancies among latent representations. However, the decoding process must be done in a strict scan order, which breaks the parallelization. We propose a parallelizable checkerboar...
Preprint
Full-text available
We present a conceptually simple, flexible and general framework for cross-dataset training in object detection. Given two or more already labeled datasets that target for different object classes, cross-dataset training aims to detect the union of the different classes, so that we do not have to label all the classes for all the datasets. By cross...
Article
Source term estimation (STE) of atmospheric dispersion plays an important role in public safety, environmental protection and many other application fields. In this paper, several new composite cost functions for STE using hybrid genetic algorithm are proposed and compared using Nemenyi test based on 68 STE tasks from Prairie Grass field experiment...
Conference Paper
Full-text available
Mass incidents have become a global problem, threatening public safety. Due to the complexity and uncertainties of mass incidents, law enforcement agencies lack analytical models and structured processes for anticipating potential threats. To address such challenge, this paper presents a threat analysis framework combining the scenario analysis met...
Article
Radiation from the deposited radionuclides is indispensable information for environmental impact assessment of nuclear power plants and emergency management during nuclear accidents. Ground shine estimation is related to multiple physical processes, including atmospheric dispersion, deposition, soil and air radiation shielding. It still remains unc...
Article
Full-text available
In recent years, terrorist attacks have become even more violent all over the world and terrorist forces have begun to spread to urban cities. Natural gas pipeline networks carrying hazardous materials in cities are easy targets for attacks. This study analyzes the vulnerability of a natural gas network to terrorist attacks. The adversary relations...
Conference Paper
Full-text available
In response to the threat of hazardous gas releases to public safety and health, we propose an agile framework for detecting and quantifying gas emission sources. Emerging techniques like high-precision gas sensors, source term estimation algorithms and Unmanned Aerial Vehicles are incorporated. The framework takes advantage of both stationary sens...
Poster
Full-text available
Explore government communication strategies on social media of Ningbo PX Incident Analyze the effects of five communication strategies: Denying, Buck-passing, Diminishing, Rebuilding, and Bolstering
Article
Full-text available
Most of the urban gas pipelines are buried underground. Small leakage exists and is difficult to detect, which is a potential threat to urban public safety. In this paper, in response to the need of leakage gas detection and assessment for buried gas network, gas dispersion in soil and atmosphere is studied in a coupled way. Different governing equ...
Article
Source term estimation for atmospheric dispersion deals with estimation of the emission strength and location of an emitting source using all available information, including site description, meteorological data, concentration observations and prior information. In this paper, Bayesian methods for source term estimation are evaluated using Prairie...
Article
Parameter estimation of a source of chemical, biological or radiological emissions is a problem of great importance for public safety. The key parameters of interest are the source intensity and its location. This paper applies the concept of Rao-Blackwell dimension reduction to solve the posterior probability distribution function of source intens...
Conference Paper
The simulation of radioactive pollutants dispersion is critical for emergency response of the nuclear terrorism. The radioactive “dirty bomb”, also called radiological dispersion device (RDD), produced and used by the terrorist to make fearful and radioactive pollution in general, has a great risk on humans. Numerical investigation of the impact of...
Conference Paper
Full-text available
In this paper, a new vulnerability metric for urban natural gas network under cascading failures are proposed, which is based on hydraulic analysis method. Simulation result of a simplified network shows that pressure-driven hydraulic method is more realistic than demand-driven hydraulic method because user’s demand is not fully satisfied under a c...
Conference Paper
Full-text available
In this study, the dispersion of natural gas leaking from buried pipelines in an urban street canyon is investigated through computer simulation. The numerical simulations were performed using OpenFOAM, a free and open source software for computational fluid dynamics (CFD). The standard solver simpleFoam was modified to include passive scalar turbu...
Conference Paper
Full-text available
Source term estimation (STE) of hazardous material (HAZMAT) releases is critical for emergency response. Such problem is usually solved with the aid of atmospheric dispersion modelling and inversion algorithms accompanied with a variety of uncertainty, including uncertainty in atmospheric dispersion models, uncertainty in meteorological data, uncer...
Conference Paper
Full-text available
It is important to forecast the pedestrian flows for organizing crowd activities and making risk assessments. In this article, the daily pedestrian flows in the Tiananmen Square are forecasted based on historical data, the distribution of holidays and weather conditions including rain, wind, temperature, relative humidity, and AQI (Air Quality Inde...

Questions

Questions (3)
Question
Hi all,
I am a phd student desperate for some  unsteady atmospheric dispersion data to validate my CFD model and data assimilation methods (include source estimation).
I have spent a lot of time in seeking data online, and I find the following datasets:
  • CODASC by Karlsruhe Institute of Technology
  • CEDVAL wind tunnel experiment by Universität Hamburg
  • Benchmark tests from Architectural Institute of Japan
  • some datasets for the Gaussian plume model like those collected by www.jsirwin.com.
Those are all fantastic datasets for steady state dispersion. However, as I have to test my data assimilation and source estimation methods, some unsteady atmospheric dispersion data are needed.
I noticed that the following three datasets are widely used for source estimation:
  • Joint Urban 2003
  • MUST
  • FUSION Field Trial 2007 (FFT07)
However, It seems that it is difficult for phd students like me who just start their research career to obtain those datasets? Actually, I have accomplished a little work about CFD modelling and source estimation for atmospheric dispersion. Now I really need the data.
Could someone give me a hand?
Regards,
Yan Wang
Question
Hi all,
I know that for urban areas with complex geometry, snappyHexMesh in OpenFOAM is a good choice.
But for urban area with many cube buildings, I want to use structured mesh. Can you give me some advice on which software to choose? 
Question
For gas dispersion in the atmosphere, gaussian plume/puff based models/softwares can be used.
But landfill gas must first migrate underground before they reach the ground surface to form an area source and begin their dispersion in the atmosphere.
Does anyone have ideas about this? Thank you in advance. (Both analytical models/softwares and numerical ones are  welcome)

Network

Cited By

Projects

Project (1)
Project
In this project, we investigate the source term estimation (STE) techniques and cognitive search strategies. The application background includes (but not limited to) hazardous gas releases in the atmosphere due to industrial accidents or terrorist attacks and fugitive emissions from natural gas pipelines.