stat441w18/Image Question Answering using CNN with Dynamic Parameter Prediction

From statwiki
Jump to navigation Jump to search

Image Question Answering using CNN with Dynamic Parameter Prediction

Presented by

Rosie Zou, Kye Wei, Glen Chalatov, Ameer Dharamshi

Introduction

Problem Setup (in words)

Previous and Related Works

As mentioned in the earlier section, one of the major goals in computer vision is to achieve holistic understanding. While relatively new interest in the computer vision community, Image Question Answering already has a growing number of researchers working on this problem.

Problem Setup (in mathematical terms)

Mathematical Background

CNNs

RNNs and GRUs

Model

VGGNet

Parameter Prediction Network

Hashing

Model Summary

Training and Results

Training

Error Reduction

Pre-trained GRUs

Fine-tuning

Results

Critique