Attention based Memory video portrait matting

03/14/2022
by   Shufeng Song, et al.
0

We proposed a novel trimap free video matting method based on the attention mechanism. By the nature of the problem, most existing approaches use either multiple computational expansive modules or complex algorithms to exploit temporal information fully. We designed a temporal aggregation module to compute the temporal coherence between the current frame and its two previous frames.

READ FULL TEXT

page 3

page 4

page 5

page 9

research
02/17/2021

Temporal Memory Attention for Video Semantic Segmentation

Video semantic segmentation requires to utilize the complex temporal rel...
research
01/05/2021

Local Memory Attention for Fast Video Semantic Segmentation

We propose a novel neural network module that transforms an existing sin...
research
06/07/2021

Video Imprint

A new unified video analytics framework (ER3) is proposed for complex ev...
research
04/26/2022

Stochastic Coherence Over Attention Trajectory For Continuous Learning In Video Streams

Devising intelligent agents able to live in an environment and learn by ...
research
03/01/2018

TSSD: Temporal Single-Shot Object Detection Based on Attention-Aware LSTM

Temporal object detection has attracted significant attention, but most ...
research
03/01/2018

TSSD: Temporal Single-Shot Detector Based on Attention and LSTM for Robotic Intelligent Perception

Temporal object detection has attracted significant attention, but most ...
research
12/07/2019

DAVID: Dual-Attentional Video Deblurring

Blind video deblurring restores sharp frames from a blurry sequence with...

Please sign up or login with your details

Forgot password? Click here to reset