A5022836568- Advanced Image Processing Techniques
- Image Enhancement Techniques
- Advanced Vision and Imaging
- Image Processing Techniques and Applications
- Image and Signal Denoising Methods
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Video Surveillance and Tracking Methods
- Advanced Image Fusion Techniques
- Visual Attention and Saliency Detection
- Optical measurement and interference techniques
- Anomaly Detection Techniques and Applications
- Image and Video Quality Assessment
- Imbalanced Data Classification Techniques
- COVID-19 diagnosis using AI
- IoT and Edge/Fog Computing
- Green IT and Sustainability
- Biomedical Text Mining and Ontologies
- Age of Information Optimization
- Pharmacovigilance and Adverse Drug Reactions
- Non-Invasive Vital Sign Monitoring
- AI in cancer detection
- Context-Aware Activity Recognition Systems
- Topic Modeling
- Digital Imaging for Blood Diseases
ETH Zurich
2017-2025
Addiction Switzerland
2021-2023
Moscow Institute of Physics and Technology
2015
Samsung (Russia)
2009
Despite a rapid rise in the quality of built-in smartphone cameras, their physical limitations - small sensor size, compact lenses and lack specific hardware, impede them to achieve results DSLR cameras. In this work we present an end-to-end deep learning approach that bridges gap by translating ordinary photos into DSLR-quality images. We propose translation function using residual convolutional neural network improves both color rendition image sharpness. Since standard mean squared loss...
Low-end and compact mobile cameras demonstrate limited photo quality mainly due to space, hardware budget constraints. In this work, we propose a deep learning solution that translates photos taken by with capabilities into DSLR-quality automatically. We tackle problem introducing weakly supervised enhancer (WESPE) - novel image-to-image Generative Adversarial Network-based architecture. The proposed model is trained under weak supervision: unlike previous works, there no need for strong...
The performance of mobile AI accelerators has been evolving rapidly in the past two years, nearly doubling with each new generation SoCs. current 4th NPUs is already approaching results CUDA-compatible Nvidia graphics cards presented not long ago, which together increased capabilities deep learning frameworks makes it possible to run complex and models on devices. In this paper, we evaluate compare all chipsets from Qualcomm, HiSilicon, Samsung, MediaTek Unisoc that are providing hardware...
As the popularity of mobile photography is growing constantly, lots efforts are being invested now into building complex hand-crafted camera ISP solutions. In this work, we demonstrate that even most sophisticated pipelines can be replaced with a single end-to-end deep learning model trained without any prior knowledge about sensor and optics used in particular device. For this, present PyNET, novel pyramidal CNN architecture designed for fine-grained image restoration implicitly learns to...
Bokeh is an important artistic effect used to highlight the main object of interest on photo by blurring all out-of-focus areas. While DSLR and system camera lenses can render this naturally, mobile cameras are unable produce shallow depth-of-field photos due a very small aperture diameter their optics. Unlike current solutions simulating bokeh applying Gaussian blur image background, in paper we propose learn realistic focus technique directly from produced cameras. For this, present...
Image super-resolution is one of the most popular computer vision problems with many important applications to mobile devices. While solutions have been proposed for this task, they are usually not optimized even common smartphone AI hardware, mention more constrained smart TV platforms that often supporting INT8 inference only. To address problem, we introduce first Mobile challenge, where target develop an end-to-end deep learning-based image can demonstrate a real-time performance on or...
Video super-resolution has recently become one of the most important mobile-related problems due to rise video communication and streaming services. While many solutions have been proposed for this task, majority them are too computationally expensive run on portable devices with limited hardware resources. To address problem, we introduce first Mobile AI challenge, where target is develop an end-to-end deep learning-based that can achieve a real-time performance mobile GPUs. The...
This paper reviews the first NTIRE challenge on perceptual image enhancement with focus proposed solutions and results. The participating teams were solving a real-world photo problem, where goal was to map low-quality photos from iPhone 3GS device same captured Canon 70D DSLR camera. considered problem embraced number of computer vision subtasks, such as denoising, resolution sharpness enhancement, color/contrast/exposure adjustment, etc. target metric used in this combined PSNR SSIM scores...
This paper reviews the first AIM challenge on mapping camera RAW to RGB images with focus proposed solutions and results. The participating teams were solving a real-world photo enhancement problem, where goal was map original low-quality from Huawei P20 device same photos captured Canon 5D DSLR camera. considered problem embraced number of computer vision subtasks, such as image demosaicing, denoising, gamma correction, resolution sharpness enhancement, etc. target metric used in this...
This paper reviews the first AIM challenge on bokeh effect synthesis with focus proposed solutions and results. The participating teams were solving a real-world image-to-image mapping problem, where goal was to map standard narrow-aperture photos same captured shallow depth-of-field by Canon 70D DSLR camera. In this task, participants had restore based only one single frame without any additional data from other cameras or sensors. target metric used in combined fidelity scores (PSNR SSIM)...
Depth estimation is an important computer vision problem with many practical applications to mobile devices. While solutions have been proposed for this task, they are usually very computationally expensive and thus not applicable on-device inference. To address problem, we introduce the first Mobile AI challenge, where target develop end-to-end deep learning-based depth that can demonstrate a nearly real-time performance on smartphones IoT platforms. For this, participants were provided new...
Image denoising is one of the most critical problems in mobile photo processing. While many solutions have been proposed for this task, they are usually working with synthetic data and too computationally expensive to run on devices. To address problem, we introduce first Mobile AI challenge, where target develop an end-to-end deep learning-based image solution that can demonstrate high efficiency smartphone GPUs. For this, participants were provided a novel large-scale dataset consisting...
As the quality of mobile cameras starts to play a crucial role in modern smartphones, more and attention is now being paid ISP algorithms used improve various perceptual aspects photos. In this Mobile AI challenge, target was develop an end-to-end deep learning-based image signal processing (ISP) pipeline that can replace classical hand-crafted ISPs achieve nearly real-time performance on smartphone NPUs. For this, participants were provided with novel learned dataset consisting RAW-RGB...
Camera scene detection is among the most popular computer vision problem on smartphones. While many custom solutions were developed for this task by phone vendors, none of designed models available publicly up until now. To address problem, we introduce first Mobile AI challenge, where target to develop quantized deep learning-based camera classification that can demonstrate a real-time performance smartphones and IoT platforms. For this, participants provided with large-scale CamSDD dataset...