Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/zhanghang1989/torch-encoding-layer
Deep Texture Encoding Network
https://github.com/zhanghang1989/torch-encoding-layer
computer-vision deep-learning deep-neural-networks encoding texture
Last synced: about 5 hours ago
JSON representation
Deep Texture Encoding Network
- Host: GitHub
- URL: https://github.com/zhanghang1989/torch-encoding-layer
- Owner: zhanghang1989
- Created: 2016-11-30T20:26:04.000Z (almost 8 years ago)
- Default Branch: master
- Last Pushed: 2020-12-21T04:01:09.000Z (almost 4 years ago)
- Last Synced: 2023-10-20T23:59:32.385Z (about 1 year ago)
- Topics: computer-vision, deep-learning, deep-neural-networks, encoding, texture
- Language: Lua
- Homepage: http://zhanghang1989.github.io/DeepEncoding/
- Size: 146 KB
- Stars: 89
- Watchers: 9
- Forks: 34
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Deep Encoding
Created by [Hang Zhang](http://hangzh.com/)
### Table of Contents
0. [Introduction](#introduction)
0. [Installation](#installation)
0. [Experiments](#experiments)
0. [Benchmarks](#benchmarks)
0. [Acknowldgements](#acknowldgements)
## Introduction
- **Please checkout our [PyTorch implementation](https://github.com/zhanghang1989/PyTorch-Encoding) (recommended, memory efficient)**.
- This repo is a Torch implementation of Encoding Layer as described in the paper:
**Deep TEN: Texture Encoding Network** [[arXiv]](https://arxiv.org/pdf/1612.02844.pdf)
[Hang Zhang](http://hangzh.com/), [Jia Xue](http://jiaxueweb.com/), [Kristin Dana](http://eceweb1.rutgers.edu/vision/dana.html)
```
@article{zhang2016deep,
title={Deep TEN: Texture Encoding Network},
author={Zhang, Hang and Xue, Jia and Dana, Kristin},
journal={arXiv preprint arXiv:1612.02844},
year={2016}
}
```
Traditional methods such as bag-of-words BoW (left) have a structural similarity to more recent FV-CNN methods (center). Each component is optimized in separate steps. In our approach (right) the entire pipeline is learned in an integrated manner, tuning each component for the task at hand (end-to-end texture/material/pattern recognition).
## Installation
On Linux
```bash
luarocks install https://raw.githubusercontent.com/zhanghang1989/Deep-Encoding/master/deep-encoding-scm-1.rockspec
```
On OSX
```bash
CC=clang CXX=clang++ luarocks install https://raw.githubusercontent.com/zhanghang1989/Deep-Encoding/master/deep-encoding-scm-1.rockspec
```
## Experiments
- The Joint Encoding experiment in Sec4.2 will execute by default (tested using 1 Titan X GPU). This achieves *12.89%* percentage error on STL-10 dataset, which is ***49.8%*** relative improvement comparing to pervious state-of-the art *25.67%* of Zhao *et. al. 2015*.:
```bash
git clone https://github.com/zhanghang1989/Deep-Encoding
cd Deep-Encoding/experiments
th main.lua
```
- Training Deep-TEN on MINC-2500 in Sec4.1 using 4 GPUs.
0. Please download the pre-trained
[ResNet-50](https://d2j0dndfm35trm.cloudfront.net/resnet-50.t7) Torch model
and the [MINC-2500](http://opensurfaces.cs.cornell.edu/static/minc/minc-2500.tar.gz) dataset to ``minc`` folder before executing the program (tested using 4 Titan X GPUs).
```bash
th main.lua -retrain resnet-50.t7 -ft true \
-netType encoding -nCodes 32 -dataset minc \
-data minc/ -nClasses 23 -batchSize 64 \
-nGPU 4 -multisize true
```
0. To get comparable results using 2 GPUs, you should change the batch size and the corresponding learning rate:
```bash
th main.lua -retrain resnet-50.t7 -ft true \
-netType encoding -nCodes 32 -dataset minc \
-data minc/ -nClasses 23 -batchSize 32 \
-nGPU 2 -multisize true -LR 0.05\
```
### Benchmarks
Dataset |MINC-2500| FMD | GTOS | KTH |4D-Light
:----------------------------|:-------:|:---:|:----:|:---:|:------:
FV-SIFT |46.0 |47.0 |65.5 |66.3 |58.4
FV-CNN(VD) |61.8 |75.0 |77.1 |71.0 |70.4
FV-CNN(VD) multi |63.1 |74.0 |79.2 |77.8 |76.5
FV-CNN(ResNet)multi|69.3 |78.2 |77.1 |78.3 |77.6
Deep-TEN\*(**ours**) |**81.3**|80.2±0.9|**84.5±2.9**|**84.5±3.5**|**81.7±1.0**
State-of-the-Art |76.0±0.2|**82.4±1.4**| 81.4|81.1±1.5|77.0±1.1
### Acknowldgements
We thank Wenhan Zhang from Physics department, Rutgers University for discussions of mathematic models.
This work was supported by National Science Foundation award IIS-1421134.
A GPU used for this research was donated by the NVIDIA Corporation.