Perceptual Loss Deep Feature Interpolation for Image Content Changes

Slides:

Advertisements

Similar presentations

Hongliang Li, Senior Member, IEEE, Linfeng Xu, Member, IEEE, and Guanghui Liu Face Hallucination via Similarity Constraints.

Advertisements

1 Expression Cloning Jung-yong Noh Ulrich Neumann Siggraph01.

Smart Traveller with Visual Translator for OCR and Face Recognition LYU0203 FYP.

Spatial Pyramid Pooling in Deep Convolutional

From R-CNN to Fast R-CNN

Face Recognition Using Neural Networks Presented By: Hadis Mohseni Leila Taghavi Atefeh Mirsafian.

Dimensionality Reduction: Principal Components Analysis Optional Reading: Smith, A Tutorial on Principal Components Analysis (linked to class webpage)

Object Bank Presenter ： Liu Changyu Advisor ： Prof. Alex Hauptmann Interest ： Multimedia Analysis April 4 th, 2013.

Mesh Watermarking based on 2D Transferred Domain Jongyun Jun Tae-Joon Kim CS548 Term Project Presentation 2010/05/13.

Fast Direct Super-Resolution by Simple Functions

Dr. Z. R. Ghassabi Spring 2015 Deep learning for Human action Recognition 1.

Feedforward semantic segmentation with zoom-out features

Foundational Issues Machine Learning 726 Simon Fraser University.

On the relevance of facial expressions for biometric recognition Marcos Faundez-Zanuy, Joan Fabregas Escola Universitària Politècnica de Mataró (Barcelona.

Parsing Natural Scenes and Natural Language with Recursive Neural Networks INTERNATIONAL CONFERENCE ON MACHINE LEARNING (ICML 2011) RICHARD SOCHER CLIFF.

Facial Smile Detection Based on Deep Learning Features Authors: Kaihao Zhang, Yongzhen Huang, Hong Wu and Liang Wang Center for Research on Intelligent.

City Forensics: Using Visual Elements to Predict Non-Visual City Attributes Sean M. Arietta, Alexei A. Efros, Ravi Ramamoorthi, Maneesh Agrawala Presented.

SUPER RESOLUTION USING NEURAL NETS Hila Levi & Eran Amar Weizmann Ins

Recent developments in object detection

Hybrid Deep Learning for Reflectance Confocal Microscopy Skin Images

Image Resampling & Interpolation

Generative Adversarial Nets

Data Mining, Neural Network and Genetic Programming

Data Mining, Neural Network and Genetic Programming

Convolutional Neural Fabrics by Shreyas Saxena, Jakob Verbeek

MR images analysis of glioma

Automatic Lung Cancer Diagnosis from CT Scans (Week 4)

Combining CNN with RNN for scene labeling (segmentation)

Part-Based Room Categorization for Household Service Robots

Figure 2 from High-Resolution Submillimeter Multiline Observations of G19.61 – 0.23: Small-Scale Chemistry (Image 1 of 2) Sheng-Li Qin et al ApJ.

Hierarchical Deep Convolutional Neural Network

Recovery from Occlusion in Deep Feature Space for Face Recognition

Sujay Yadawadkar, Virginia Tech

Adversarially Tuned Scene Generation

Image Question Answering

Low Dose CT Image Denoising Using WGAN and Perceptual Loss

By: Kevin Yu Ph.D. in Computer Engineering

Image Classification.

PixelGAN Autoencoders

Convolutional Neural Networks for Visual Tracking

Deep Learning Tutorial

Towards Understanding the Invertibility of Convolutional Neural Networks Anna C. Gilbert1, Yi Zhang1, Kibok Lee1, Yuting Zhang1, Honglak Lee1,2 1University.

Vessel Extraction in X-Ray Angiograms Using Deep Learning

RGB-D Image for Scene Recognition by Jiaqi Guo

Presented by Shixing Chen

Super-Resolution-based inpainting

GPU Accelerated Image Super-Resolution

Non-Stationary Texture Synthesis by Adversarial Expansion

Word embeddings based mapping

Word embeddings based mapping

Lip movement Synthesis from Text

RCNN, Fast-RCNN, Faster-RCNN

Artifacts of Adversarial Examples

Outline Announcement Perceptual organization, grouping, and segmentation Hough transform Read Chapter 17 of the textbook File: week14-m.ppt.

Image Resampling & Interpolation

Machine learning overview

Zeroshot Learning Mun Jonghwan.

TPGAN overview.

Unsupervised Perceptual Rewards For Imitation Learning

Department of Computer Science Ben-Gurion University of the Negev

Learning and Memorization

Unrolling the shutter: CNN to correct motion distortions

Task Produce artistic portrait drawings.. Task Produce artistic portrait drawings.

End-to-End Facial Alignment and Recognition

Text-to-speech (TTS) Traditional approaches (before 2016) Neural TTS

Single image super-resolution with limited number of filters

Recent Developments on Super-Resolution

Directional Occlusion with Neural Network

ICLR, 2019 Jiahe Li

Presentation transcript:

Perceptual Loss Deep Feature Interpolation for Image Content Changes Upchurch P, Gardner J, Bala K, et al. arXiv 2016. Perceptual Losses for Real-Time Style Transfer and Super-Resolution Johnson J, Alahi A, Fei-Fei L. ECCV 2016 citation:43

Outline What’s Perceptual Loss Image Content Changes and Style Transfer Style Transfer Real-Time Style Transfer Features Interpolation for Image Content Changes

Outline What’s Perceptual Loss Image Content Changes and Style Transfer Style Transfer Real-Time Style Transfer Features Interpolation for Image Content Changes

What’s Perceptual Loss Perceptual Loss : differences between high-level image feature presentations extracted from pre-trained CNN

What’s Perceptual Loss Perceptual Loss : differences between high-level image feature presentations extracted from pre-trained CNN groundtruth MSE: 0.0097 MSE: 0.0097

Pixel-level Loss failed to capture perceptual differences! What’s Perceptual Loss Perceptual Loss : differences between high-level image feature presentations extracted from pre-trained CNN Pixel-level Loss failed to capture perceptual differences! groundtruth MSE: 0.0097 MSE: 0.0097

Outline What’s Perceptual Loss Image Content Changes and Style Transfer Style Transfer Real-Time Style Transfer Features Interpolation for Image Content Changes

Outline What’s Perceptual Loss Image Content Changes and Style Transfer Style Transfer Real-Time Style Transfer Features Interpolation for Image Content Changes

Style Transfer

Content loss:

Content loss: Style loss:

Content loss: Style loss: Total loss:

Outline What’s Perceptual Loss Image Content Changes and Style Transfer Style Transfer Real-Time Style Transfer Features Interpolation for Image Content Changes

Outline What’s Perceptual Loss Image Content Changes and Style Transfer Style Transfer Real-Time Style Transfer Features Interpolation for Image Content Changes

Content loss:

Content loss: Style loss:

Content loss: Style loss: Total variation loss: loss

Content loss: Style loss: Total variation loss: loss Total:

Outline What’s Perceptual Loss Image Content Changes and Style Transfer Style Transfer Real-Time Style Transfer Features Interpolation for Image Content Changes

Outline What’s Perceptual Loss Image Content Changes and Style Transfer Style Transfer Real-Time Style Transfer Features Interpolation for Image Content Changes

Aging: Facial Attribute: Deep Feature Interpolation for Image Content Changes Aging: Facial Attribute:

This demonstration of unsupervised generative models learning object attributes like scale, rotation, position, and semantics was one of the first.

Step 1: Map target and source images into deep feature space:

Step 1: Map target and source images into deep feature space: Step 2: Compute attribute vector:

Step 1: Map target and source images into deep feature space: Step 2: Compute attribute vector: Step 3: Interpolate test image:

Step 1: Map target and source images into deep feature space: Step 2: Compute attribute vector: Step 3: Interpolate test image: Step 4: Reverse mapping:

Reverse mapping:

Thank you