Skip to content
master
Go to file
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
tx1
 
 
 
 
 
 
 
 

README.md

Pedestrian-Detection-on-TX1

##Introduction We compress the pedestrian detection model from ResNet-200 (63 millions parameter) to our fixed channel ResNet-18 (0.157 million parameter). Our paper utilizes the idea of Knowledge Distillation with extra helps from model confidence and hint layer to achieve 400x compression with 4.9% log-average miss rate drop. For more detail, please refer to our arXiv paper and slides.

pipeline

##Result Log-average miss rate on Caltech (lower is better)

Model Log-avg MR #Parameters Time (Titan X) Memory
ResNet-200 17.5% 63M 24ms 5377MB
ResNet-18 18.0% 11M 3ms 937MB
ResNet-18-Thin 20.3% 2.8M 3ms 633MB
ResNet-18-Small 22.4% 0.157M 3ms 565MB
*Results are from the highest improvement method (Hint+Conf).

##Demo

pipeline

##Installation ###Training The networks were trained by torch-nnet-trainer. Please set up caltech10x dataset according to Hosang.

###Testing on TX1 Ideally, the dynamic library should work for TX1. However, if an issue occurs, please build SquareChnnlFltrs for region proposal and replace "libmonocular_objects_detection.so".

The input is a set of images from video (extract by convert.sh). Use "make forward" for building forward.cpp, and "make run --input($input) --output($output)" to forward the images.

##Citing our model If you found our model useful, please cite our paper:

@articles{ShenTrust2016,
  title = {In Teacher We Trust: Learning Compressed Models for Pedestrian Detection},
  author = {Shen, J., Vesdapunt, N., Boddeti, V.~N., Kitani, K.~M. and Osterwood, C.},
  journal = {arXiv preprint, arXiv:1612.00478},
  year = {2016}
}

Releases

No releases published

Packages

No packages published

Languages

You can’t perform that action at this time.