YaoAnbang / SAN

[ECCV 2020] Learning to Learn Parameterized Classification Networks for Scalable Input Images

Scale Adaptive Network

Official implementation of Scale Adaptive Network (SAN) as described in Learning to Learn Parameterized Classification Networks for Scalable Input Images (ECCV'20) by Duo Li, Anbang Yao and Qifeng Chen on the ILSVRC 2012 benchmark.

We present a meta learning framework which dynamically parameterizes main networks conditioned on its input resolution at runtime, leading to efficient and flexible inference for arbitrarily switchable input resolutions.

Requirements

Dependency

• PyTorch 1.0+
• NVIDIA-DALI (in development, not recommended)

Dataset

Download the ImageNet dataset and move validation images to labeled subfolders. To do this, you can use the following script: https://raw.githubusercontent.com/soumith/imagenetloader.torch/master/valprep.sh

Pre-trained Models

Baseline (individually trained on each resolution)

ResNet-18

Resolution	Top-1 Acc.	Download
224x224	70.974	link
192x192	69.754	link
160x160	68.482	link
128x128	66.360	link
96x96	62.560	link

ResNet-50

Resolution	Top-1 Acc.	Download
224x224	77.150	link
192x192	76.406	link
160x160	75.312	link
128x128	73.526	link
96x96	70.610	link

MobileNetV2

Please visit my repository mobilenetv2.pytorch.

SAN

Architecture	Download
ResNet-18	link
ResNet-50	link
MobileNetV2	link

Training

ResNet-18/50

python imagenet.py \
    -a meta_resnet18/50 \
    -d <path-to-ILSVRC2012-data> \
    --epochs 120 \
    --lr-decay cos \
    -c <path-to-save-checkpoints> \
    --sizes <list-of-input-resolutions> \ # default is 224, 192, 160, 128, 96
    -j <num-workers>
    --kd

MobileNetV2

python imagenet.py \
    -a meta_mobilenetv2 \
    -d <path-to-ILSVRC2012-data> \
    --epochs 150 \
    --lr-decay cos \
    --lr 0.05 \
    --wd 4e-5 \
    -c <path-to-save-checkpoints> \
    --sizes <list-of-input-resolutions> \ # default is 224, 192, 160, 128, 96
    -j <num-workers>
    --kd

Testing

Proxy Inference (default)

python imagenet.py \
    -a <arch> \
    -d <path-to-ILSVRC2012-data> \
    --resume <checkpoint-file> \
    --sizes <list-of-input-resolutions> \
    -e
    -j <num-workers>

Arguments are:

• checkpoint-file: previously downloaded checkpoint file from here.
• list-of-input-resolutions: test resolutions using different privatized BNs.

which gives Table 1 in the main paper and Table 5 in the supplementary materials.

Ideal Inference

Manually set the scale encoding here, which gives the left panel of Table 2 in the main paper.

Uncomment this line in the main script to enable post-hoc BN calibration, which gives the middle panel of Table 2 in the main paper.

Data-Free Ideal Inference

Manually set the scale encoding here and its corresponding shift here, then uncomment this line to replace its above line, which gives Table 6 in the supplementary materials.

Citation

If you find our work useful in your research, please consider citing:

@InProceedings{Li_2020_ECCV,
author = {Li, Duo and Yao, Anbang and Chen, Qifeng},
title = {Learning to Learn Parameterized Classification Networks for Scalable Input Images},
booktitle = {The European Conference on Computer Vision (ECCV)},
month = {August},
year = {2020}
}