We conduct research on cognitive and cybernetic vision systems to automate processes. The research activity of the group includes developing models for visual information processing and the development of algorithms that learn in a supervised and self-supervised manner from continuous and limited data as well as from clean and noisy data. Our work is firmly grounded in applications. The lab focuses on transferring the developed technologies to innovative single and multi-camera systems, aided by additional modalities, actuators, and efficient cloud and edge computing.
The project aims at developing an innovative, automated wide-field, high-resolution, multi-camera system for the optical inspection of wind turbines. The computing engine behind the inspection system will take advantage of multi-task video processing at the edge.
The project aims at developing an innovative, automated wide-field, high-resolution, multi-camera system for the optical inspection of wind turbines. The computing engine behind the inspection system will take advantage of multi-task video processing at the edge.
Scientists from the ETI Faculty of Gdańsk University of Technology are participating in the work on creating a device that works with commonly used dental tips for non-invasive, optical control of the inside of the oral cavity when performing a dental procedure. The project is carried out by Master Level Technologies in cooperation with the GUT special purpose vehicle Excento, as well as Bibus Menos and Lambda System.
Scientists from the ETI Faculty of Gdańsk University of Technology are participating in the work on creating a device that works with commonly used dental tips for non-invasive, optical control of the inside of the oral cavity when performing a dental procedure. The project is carried out by Master Level Technologies in cooperation with the GUT special purpose vehicle Excento, as well as Bibus Menos and Lambda System.
We introduce a new, asymmetrically annotated dataset of natural teeth in phantom scenes for multi-task video processing: restoration, teeth segmentation, and inter-frame homography estimation. Pairs of frames were acquired with a beam splitter. The dataset constitutes a low-quality frame, its high-quality counterpart, a teeth segmentation mask, and an inter-frame homography matrix.
We introduce a new, asymmetrically annotated dataset of natural teeth in phantom scenes for multi-task video processing: restoration, teeth segmentation, and inter-frame homography estimation. Pairs of frames were acquired with a beam splitter. The dataset constitutes a low-quality frame, its high-quality counterpart, a teeth segmentation mask, and an inter-frame homography matrix.
BP-EVD: Forward Block-Output Propagation for Efficient Video Denoising
We conduct research on cognitive and cybernetic vision systems to automate processes. The research activity of the group includes developing models for visual information processing and the development of algorithms that learn in a supervised and self-supervised manner from continuous and limited data as well as from clean and noisy data. Our work is firmly grounded in applications. The lab focuses on transferring the developed technologies to innovative single and multi-camera systems, aided by additional modalities, actuators, and efficient cloud and edge computing.
Multi-task Video Enhancement for Dental Interventions
A microcamera firmly attached to a dental handpiece allows dentists to continuously monitor the progress of conservative dental procedures. Video enhancement in video-assisted dental interventions alleviates low-light, noise, blur, and camera handshakes that collectively degrade visual comfort. To this end, we introduce a novel deep network for multi-task video enhancement that enables macro-visualization of dental scenes. In particular, the proposed network jointly leverages video restoration and temporal alignment in a multi-scale manner for effective video enhancement.
Concurrent video denoising and deblurring for dynamic scenes
Dynamic scene video deblurring is a challenging task due to the spatially variant blur inflicted by independently moving objects and camera shakes. This is a difficult yet simplified scenario because noise is neglected when it is omnipresent in a wide spectrum of video processing applications. Despite its relevance, the problem of concurrent noise and dynamic blur has not yet been addressed in the deep learning literature. To this end, we analyze existing state-of-the-art deblurring methods and encounter their limitations in handling non-uniform blur under strong noise conditions.
Cameras and Algorithms Lab
Our group conducts basic research on cognitive and cybernetic vision systems to automate processes. We develop models for visual information processing and supervised and self-supervised machine learning algorithms that train the models on temporal, continuous, scarce, clean, and noisy data. Our work is firmly grounded in applications. The lab focuses on transferring the developed technologies to innovative single and multi-camera systems, aided by additional modalities, actuators, and efficient cloud and edge computing.
The project aims at developing an innovative, automated wide-field, high-resolution, multi-camera system for the optical inspection of wind turbines. The computing engine behind the inspection system will take advantage of multi-task video processing at the edge.
The project aims at developing an innovative, automated wide-field, high-resolution, multi-camera system for the optical inspection of wind turbines. The computing engine behind the inspection system will take advantage of multi-task video processing at the edge.
Scientists from the ETI Faculty of Gdańsk University of Technology are participating in the work on creating a device that works with commonly used dental tips for non-invasive, optical control of the inside of the oral cavity when performing a dental procedure. The project is carried out by Master Level Technologies in cooperation with the GUT special purpose vehicle Excento, as well as Bibus Menos and Lambda System.
Scientists from the ETI Faculty of Gdańsk University of Technology are participating in the work on creating a device that works with commonly used dental tips for non-invasive, optical control of the inside of the oral cavity when performing a dental procedure. The project is carried out by Master Level Technologies in cooperation with the GUT special purpose vehicle Excento, as well as Bibus Menos and Lambda System.
We introduce a new, asymmetrically annotated dataset of natural teeth in phantom scenes for multi-task video processing: restoration, teeth segmentation, and inter-frame homography estimation. Pairs of frames were acquired with a beam splitter. The dataset constitutes a low-quality frame, its high-quality counterpart, a teeth segmentation mask, and an inter-frame homography matrix.
We introduce a new, asymmetrically annotated dataset of natural teeth in phantom scenes for multi-task video processing: restoration, teeth segmentation, and inter-frame homography estimation. Pairs of frames were acquired with a beam splitter. The dataset constitutes a low-quality frame, its high-quality counterpart, a teeth segmentation mask, and an inter-frame homography matrix.
Many deep learning applications require figure-ground segmentation. The performance of segmentation models varies across modalities and acquisition settings. The release of SegSperm dataset intends to contribute to the development of algorithms for: (1) segmentation of small objects, (2) segmentation of blurred objects, (3) Computer-Assisted Sperm Analysis (CASA) systems.
We introduce a new, asymmetrically annotated dataset of natural teeth in phantom scenes for multi-task video processing: restoration, teeth segmentation, and inter-frame homography estimation. Pairs of frames were acquired with a beam splitter. The dataset constitutes a low-quality frame, its high-quality counterpart, a teeth segmentation mask, and an inter-frame homography matrix.
BP-EVD: Forward Block-Output Propagation for Efficient Video Denoising
Denoising videos in real-time is critical in many applications, including robotics and medicine, where varying-light conditions, miniaturized sensors, and optics can substantially compromise image quality. This work proposes the first video denoising method based on a deep neural network that achieves state-of-the-art performance on dynamic scenes while running in real-time on VGA video resolution with no frame latency. The backbone of our method is a novel, remarkably simple, temporal network of cascaded blocks with forward block output propagation. We train our architecture with short, long, and global residual connections by minimizing the restoration loss of pairs of frames, leading to a more effective training across noise levels. It is robust to heavy noise following Poisson-Gaussian noise statistics. The algorithm is evaluated on RAW and RGB data. We propose a denoising algorithm that requires no future frames to denoise a current frame, reducing its latency considerably. The visual and quantitative results show that our algorithm achieves state-of-the-art performance among efficient algorithms, achieving from two-fold to two-orders-of-magnitude speed-ups on standard benchmarks for video denoising.
Multi-task Video Enhancement for Dental Interventions
A microcamera firmly attached to a dental handpiece allows dentists to continuously monitor the progress of conservative dental procedures. Video enhancement in video-assisted dental interventions alleviates low-light, noise, blur, and camera handshakes that collectively degrade visual comfort. To this end, we introduce a novel deep network for multi-task video enhancement that enables macro-visualization of dental scenes. In particular, the proposed network jointly leverages video restoration and temporal alignment in a multi-scale manner for effective video enhancement.
Concurrent video denoising and deblurring for dynamic scenes
Dynamic scene video deblurring is a challenging task due to the spatially variant blur inflicted by independently moving objects and camera shakes. This is a difficult yet simplified scenario because noise is neglected when it is omnipresent in a wide spectrum of video processing applications. Despite its relevance, the problem of concurrent noise and dynamic blur has not yet been addressed in the deep learning literature. To this end, we analyze existing state-of-the-art deblurring methods and encounter their limitations in handling non-uniform blur under strong noise conditions.
