Track topics on Twitter Track topics that are important to you
In this paper, we propose a design scheme for deep learning networks in face parsing task with promising accuracy and real-time inference speed. By analyzing the differences between general image parsing task and face parsing task, we first revisit the structure of traditional FCN and make improvements to adapt to the unique properties of face parsing task. Especially, the concept of Normalized Receptive Field is proposed to give more insights on designing the network. Then a novel loss function called Statistical Contextual Loss is introduced, which integrates richer contextual information and regularizes features during training. For further model acceleration, we propose a semi-supervised distillation scheme that effectively transfers the learned knowledge to a lighter network. Extensive experiments on LFW and Helen dataset demonstrate the significant superiority of the new design scheme on both efficacy and efficiency.
This article was published in the following journal.
Name: IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
Various biosensing platforms for real-time monitoring and mapping of chemical signals in neural networks have been developed based on CMOS process technology. Despite their achievements, however, ther...
The aim of study is to evaluate the general performance and efficiency of the using real time intraoperative ultrasound system with Volume Navigation system technology in glioma. Compare glioma intrao...
In medical imaging, the availability of robust and accurate automatic segmentation methods is very important for a user-independent and time-saving delineation of regions of interest. In this work, we...
Faces play important roles in the social lives of humans. Besides real faces, people also encounter numerous cartoon faces in daily life which convey basic emotional states through facial expressions....
We implemented a real-time report to distribute respiratory pathogen data for our 8-hospital system to anyone with an Internet connection and a web browser. Real-time access to accurate regional labor...
Quantitative measurements of left ventricular volume and ejection fraction are useful in the management of patients with heart disease. Several imaging methods exist, but are limited by c...
This is a prospective randomized clinical trial examining how IRIS (Intelligent Real-time Image Segmentation) affects biopsy patterns in VLE (Volumetric laser endomicroscopy).
This project focuses on the further development and clinical testing of an image-guided surgical system. The system will help surgeons perform procedures that involve inserting a screw, gu...
The goal is to achieve the maximal radiotherapy tumor dose while sparing the health tissue and critical structures. On-board cone-beam CT (CBCT) scans are routinely acquired prior to dose ...
The purpose of this study is to determine whether facial exercises in conjunction with opening exercises routinely provided after facial surgery to correct a facial skeletal disharmony wil...
Ultrasonography applying the Doppler effect combined with real-time imaging. The real-time image is created by rapid movement of the ultrasound beam. A powerful advantage of this technique is the ability to estimate the velocity of flow from the Doppler shift frequency.
A visual image which is recalled in accurate detail. It is a sort of projection of an image on a mental screen.
Echocardiography applying the Doppler effect, with the superposition of flow information as colors on a gray scale in a real-time image.
Surgical procedures conducted with the aid of computers. This is most frequently used in orthopedic and laparoscopic surgery for implant placement and instrument guidance. Image-guided surgery interactively combines prior CT scans or MRI images with real-time video.
Endoscopic surgical procedures performed with visualization via video transmission. When real-time video is combined interactively with prior CT scans or MRI images, this is called image-guided surgery (see SURGERY, COMPUTER-ASSISTED).