Compact CNN for Indexing Egocentric Videos

04/28/2015
by   Yair Poleg, et al.
0

While egocentric video is becoming increasingly popular, browsing it is very difficult. In this paper we present a compact 3D Convolutional Neural Network (CNN) architecture for long-term activity recognition in egocentric videos. Recognizing long-term activities enables us to temporally segment (index) long and unstructured egocentric videos. Existing methods for this task are based on hand tuned features derived from visible objects, location of hands, as well as optical flow. Given a sparse optical flow volume as input, our CNN classifies the camera wearer's activity. We obtain classification accuracy of 89 the current state-of-the-art by 19 extended egocentric video dataset, classifying twice the amount of categories than current state-of-the-art. Furthermore, our CNN is able to recognize whether a video is egocentric or not with 99.2 current state-of-the-art. To better understand what the network actually learns, we propose a novel visualization of CNN kernels as flow fields.

READ FULL TEXT

page 1

page 2

page 4

page 6

page 8

research
04/06/2020

Cascaded Deep Video Deblurring Using Temporal Sharpness Prior

We present a simple and effective deep convolutional neural network (CNN...
research
05/04/2019

Learning Spatio-Temporal Features with Two-Stream Deep 3D CNNs for Lipreading

We focus on the word-level visual lipreading, which requires recognizing...
research
01/27/2023

Optical Flow Estimation in 360^∘ Videos: Dataset, Model and Application

Optical flow estimation has been a long-lasting and fundamental problem ...
research
12/12/2017

Im2Flow: Motion Hallucination from Static Images for Action Recognition

Existing methods to recognize actions in static images take the images a...
research
12/07/2022

DroneAttention: Sparse Weighted Temporal Attention for Drone-Camera Based Activity Recognition

Human activity recognition (HAR) using drone-mounted cameras has attract...
research
02/19/2018

Learning Representative Temporal Features for Action Recognition

In this paper we present a novel video classification methodology that a...
research
04/08/2015

Evaluating Two-Stream CNN for Video Classification

Videos contain very rich semantic information. Traditional hand-crafted ...

Please sign up or login with your details

Forgot password? Click here to reset