Video Enhancement Based on Unpaired Learning

2021 ◽  
Author(s):  
Jinjin Chen ◽  
Wenyao Gan ◽  
Hengsheng Zhang ◽  
Rong Xie ◽  
Li Song ◽  
...  
Keyword(s):  
Video Enhancement

Modulated polarization microscopy displays the dynamics of cytoskeletal structures in living, motile cells

1995 ◽  
Vol 53 ◽  
pp. 902-903
Author(s):  
J. R. Kuhn ◽  
M. Poenie
Keyword(s):  
Mitotic Spindle ◽  
Actin Filaments ◽  
Cell Shape ◽  
Dynamic Structure ◽  
Living Cells ◽  
Cytoskeletal Proteins ◽  
Polarization Microscopy ◽  
Video Enhancement ◽  
Ultrastructural Studies ◽  
Motile Cells

Cell shape and movement are controlled by elements of the cytoskeleton including actin filaments an microtubules. Unfortunately, it is difficult to visualize the cytoskeleton in living cells and hence follow it dynamics. Immunofluorescence and ultrastructural studies of fixed cells while providing clear images of the cytoskeleton, give only a static picture of this dynamic structure. Microinjection of fluorescently Is beled cytoskeletal proteins has proved useful as a way to follow some cytoskeletal events, but long terry studies are generally limited by the bleaching of fluorophores and presence of unassembled monomers.Polarization microscopy has the potential for visualizing the cytoskeleton. Although at present, it ha mainly been used for visualizing the mitotic spindle. Polarization microscopy is attractive in that it pro vides a way to selectively image structures such as cytoskeletal filaments that are birefringent. By combing ing standard polarization microscopy with video enhancement techniques it has been possible to image single filaments. In this case, however, filament intensity depends on the orientation of the polarizer and analyzer with respect to the specimen.

Automated Ultrasound Video Enhancement

2021 ◽  
Author(s):  
Shaima I. Jabbar ◽  
Charles R. Day ◽  
Abathar Q. Aladi ◽  
Edward K. Chadwick
Keyword(s):  
Video Enhancement

Exploring the Effect of Immediate Video Feedback on Coaching

2016 ◽  
Vol 32 (1) ◽  
pp. 47-53 ◽  
Author(s):  
Jessica Suhrheinrich ◽  
Janice Chan
Keyword(s):  
Special Education Teachers ◽  
Univariate Analysis ◽  
Autism Spectrum ◽  
Video Feedback ◽  
Video Enhancement ◽  
Significant Group ◽  
Evidence Based Practices ◽  
Training Outcomes ◽  
Participant Satisfaction ◽  
Service Training

Although evidence-based practices for autism spectrum disorders exist, they are often not effectively incorporated into school-based programs, indicating a need for enhanced training strategies for educators. This study examined the effects of immediate video feedback during coaching for teachers and paraprofessionals learning Classroom Pivotal Response Teaching (CPRT). Special education teachers, along with their classroom paraprofessionals, were randomly assigned to a coaching as usual (CAU) or a coaching with video enhancement (VE) condition. Both groups received both verbal and written feedback regarding strengths and weaknesses of their CPRT implementation. Additionally, the VE condition received video feedback during their coaching sessions. Overall, teachers demonstrated higher fidelity of implementation than paraprofessionals, t(44) = −2.73, p < .01, but no significant group differences were identified between VE and CAU conditions. Univariate analysis of variance models were conducted to examine the relationship between participant satisfaction regarding overall quality of the training and highest percentage of CPRT components passed, F(2, 37) = 3.93, p = .03. Results indicate use of the iPad may impact training outcomes and participant satisfaction with training procedures and add to the very limited literature on how technology may be used to enhance in-service training for teachers.

scholarly journals Spatio-temporal deep learning model for distortion classification in laparoscopic video

F1000Research ◽  
2021 ◽  
Vol 10 ◽  
pp. 1010
Author(s):  
Nouar AlDahoul ◽  
Hezerul Abdul Karim ◽  
Abdulaziz Saleh Ba Wazir ◽  
Myles Joshua Toledo Tan ◽  
Mohammad Faizal Ahmad Fauzi
Keyword(s):  
Deep Learning ◽  
Large Scale ◽  
Spatial Information ◽  
Short Term Memory ◽  
Learning Model ◽  
Motion Blur ◽  
Video Frame ◽  
Video Enhancement ◽  
Deep Learning Model ◽  
Laparoscopic Videos

Background: Laparoscopy is a surgery performed in the abdomen without making large incisions in the skin and with the aid of a video camera, resulting in laparoscopic videos. The laparoscopic video is prone to various distortions such as noise, smoke, uneven illumination, defocus blur, and motion blur. One of the main components in the feedback loop of video enhancement systems is distortion identification, which automatically classifies the distortions affecting the videos and selects the video enhancement algorithm accordingly. This paper aims to address the laparoscopic video distortion identification problem by developing fast and accurate multi-label distortion classification using a deep learning model. Current deep learning solutions based on convolutional neural networks (CNNs) can address laparoscopic video distortion classification, but they learn only spatial information. Methods: In this paper, utilization of both spatial and temporal features in a CNN-long short-term memory (CNN-LSTM) model is proposed as a novel solution to enhance the classification. First, pre-trained ResNet50 CNN was used to extract spatial features from each video frame by transferring representation from large-scale natural images to laparoscopic images. Next, LSTM was utilized to consider the temporal relation between the features extracted from the laparoscopic video frames to produce multi-label categories. A novel laparoscopic video dataset proposed in the ICIP2020 challenge was used for training and evaluation of the proposed method. Results: The experiments conducted show that the proposed CNN-LSTM outperforms the existing solutions in terms of accuracy (85%), and F1-score (94.2%). Additionally, the proposed distortion identification model is able to run in real-time with low inference time (0.15 sec). Conclusions: The proposed CNN-LSTM model is a feasible solution to be utilized in laparoscopic videos for distortion identification.

Sign in / Sign up

Export Citation Format

Share Document