A5100371411- Advanced Vision and Imaging
- Image Enhancement Techniques
- Photovoltaic System Optimization Techniques
- Advanced Image and Video Retrieval Techniques
- Optical measurement and interference techniques
- Solar Radiation and Photovoltaics
- Image Processing Techniques and Applications
- Computer Graphics and Visualization Techniques
- Color Science and Applications
- Remote Sensing and LiDAR Applications
- Innovative Teaching and Learning Methods
- 3D Surveying and Cultural Heritage
- Microbial Community Ecology and Physiology
- Marine and coastal ecosystems
- Problem and Project Based Learning
- Communication in Education and Healthcare
- Water Treatment and Disinfection
- Innovations in Medical Education
- Water Quality Monitoring and Analysis
- Online and Blended Learning
- Generative Adversarial Networks and Image Synthesis
- Aquatic Ecosystems and Phytoplankton Dynamics
- Remote Sensing and Land Use
- Odor and Emission Control Technologies
Université Laval
2015-2023
University Town of Shenzhen
2018
Harbin Institute of Technology
2018
Guangzhou University
2018
Qiqihar University
2017
Beihang University
2015
Fourth Affiliated Hospital of China Medical University
2014
China Medical University
2014
Outdoor lighting has extremely high dynamic range. This makes the process of capturing outdoor environment maps notoriously challenging since special equipment must be used. In this work, we propose an alternative approach. We first capture with a regular, LDR omnidirectional camera, and aim to recover HDR after fact via novel, learning-based inverse tonemapping method. deep autoencoder framework which regresses linear, range data from non-linear, saturated, low panoramas. validate our...
We present a neural network that predicts HDR outdoor illumination from single LDR image. At the heart of our work is method to accurately learn lighting panoramas under any weather condition. achieve this by training another CNN (on combination synthetic and real images) take as input an panorama, regress parameters Lalonde-Mathews model. This model trained such it a) reconstructs appearance sky, b) renders objects lit illumination. use label large-scale dataset with them train image...
Identifying typical odor-causing compounds is essential for odor problem control in drinking water. In this study, aiming at a major water source reservoir hot and humid areas southern China, which encountered seasonable problems recent years, an integrated approach including comprehensive two-dimensional gas chromatography with time-of-flight mass spectrometry (GC × GC–TOFMS), flavor profile analysis (FPA) quantitative real-time polymerase chain reaction (qPCR) was adopted to investigate...
We present a neural network that predicts HDR outdoor illumination from single LDR image. At the heart of our work is method to accurately learn lighting panoramas under any weather condition. achieve this by training another CNN (on combination synthetic and real images) take as input an panorama, regress parameters Lalonde-Matthews model. This model trained such it a) reconstructs appearance sky, b) renders objects lit illumination. use label large-scale dataset with them train image...
Photometric Stereo has been explored extensively in laboratory conditions since its inception. Recently, attempts have made at applying this technique under natural outdoor lighting. Outdoor photometric stereo presents additional challenges as one does not control over illumination anymore. In paper, we explore the stability of surface normals reconstructed outdoors. We present a data-driven analysis based on large database HDR environment maps. Given sequence object images and corresponding...
While Photometric Stereo (PS) has long been confined to the lab, there a recent interest in applying this technique reconstruct outdoor objects and scenes. Un-fortunately, most successful PS techniques typically require gathering either months of data, or waiting for particular time year. In paper, we analyze illumination requirements single-day yield stable normal reconstructions, determine that these are often available much less than full day. particular, show right set conditions...
The results indicate that the main biofilm communities in different pipe materials are significantly from each other. With passage of time, richness and diversity microbial community cast iron shows a downwards trend.
Outdoor lighting has extremely high dynamic range. This makes the process of capturing outdoor environment maps notoriously challenging since special equipment must be used. In this work, we propose an alternative approach. We first capture with a regular, LDR omnidirectional camera, and aim to recover HDR after fact via novel, learning-based inverse tonemapping method. deep autoencoder framework which regresses linear, range data from non-linear, saturated, low panoramas. validate our...
Compositing virtual objects into real background images requires one to carefully match the scene's camera parameters, surface geometry, textures, and lighting obtain plausible renderings. Recent learning approaches have shown many scene properties can be estimated from images, resulting in robust automatic single-image compositing systems, but challenges remain. In particular, interactions between synthetic shadows are not handled gracefully by existing methods, which typically assume a...
Most computer vision research focuses on narrow angle lenses and is not adapted to super-wide-angle (aka spherical) lenses. This mainly because current neural networks are designed or trained interpret the significant barrel distortion that introduced in captured image by such wide As these capture a half-sphere section of sphere object space, appears when projected 2D flat sensor. By controlling this at lens design stage, camera designers can create some areas with augmented resolution...
Predicting the short-term power output of a photovoltaic panel is an important task for efficient management smart grids. Short-term forecasting at minute scale, also known as nowcasting, can benefit from sky images captured by regular cameras and installed close to solar panel. However, estimating weather conditions these images---sun intensity, cloud appearance movement, etc.---is very challenging that community has yet solve with traditional computer vision techniques. In this work, we...
We developed a Convolutional Neural Network to estimate depth on wide-angle images using panomorph lens with controlled distortion. simulated three different model and compared their performances based zone of augmented resolution.