text
stringlengths 0
820
|
---|
224px, .7m GSD |
Ground Truth Image Ihr |
(448px, .3m GSD)resampleMAE Encoder |
Transformers |
14 14GSDPE |
L1 loss |
448224 |
Scale-MAE Decoder |
L2 loss |
Loss |
Upsampling |
Reconstruction |
28 |
56 |
Demask |
High freq. |
Ground Truth |
LB |
Low freq. |
Ground TruthDeconv |
DeconvLB |
Ground Truth Image I hr |
(448px, .3m GSD) |
DecodingFigure 2. Scale-MAE employs the Masked Autoencoder framework. An input image is patchified and masked before being passed into an |
MAE encoder. A Ground Sample Distance Positional Encoding (GSDPE) is added to the encoder input, which scales the positional encodings |
to the area of ground covered. The Scale-MAE decoders has three stages: (1) Decoding, which uses a smaller number of transformer layers |
than MAE to decode the encoded values (2) Upsampling, which progressively deconvolves the decoded feature map to a larger size before |
being passed through the Laplacian Blocks (abbreviated LB, see Section 3), (3) Reconstruction, which then reconstructs low and high |
frequency features at different scales. These outputs are used to compute an aggregate loss with ground truth low and high frequency features, |
where following super resolution literature [2], an L1 loss is used for high frequency output to better reconstruct edges and an L2 loss is used |
for low frequency output to better reconstruct average values. |
data at a specific scale [13, 20, 22, 32, 41]. In this paper we |
present Scale-MAE , a masked reconstruction model that ex- |
plicitly learns relationships between data at different, known |
scales throughout the pretraining process. By leveraging this |
information, Scale-MAE produces a pretrained model that |
performs better across a wide range of GSDs and tasks. |
Masked Autoencoders [26] offer self-supervised learn- |
ing without explicit augmentations. A standard Masked |
Autoencoder resizes/crops an image, masks the majority of |
the transformed image, and then tasks a Vision Transformer |
(ViT) based autoencoder with embedding the unmasked com- |
ponents. A decoding ViT then decodes the full image from |
these learned embeddings, where the decoder is later dis- |
carded and the encoder is used to produce representations |
for an unmasked input image. |
Existing MAE-based pretraining approaches fail to gen- |
eralize across domains with images at multiple scales. |
Scale-MAE (Figure 1) overcomes this through a GSD-based |
positional encoding derived from the land area covered in the |
image. This informs the ViT of both the position and scale of |
the input image. Scale-MAE also uses a Laplacian-pyramid |
decoder to encourage the network to learn multiscale rep- |
resentations. The embeddings are decoded to two images |
containing low and residual high frequency information, re- |
spectively – see Figure 2. As we discuss in Section 3, this |
structure allows the ViT decoder to use fewer parameters |
than MAE while still producing strong representations across |
multiple scales. |
We show that Scale-MAE leads to better performing, |
more robust multiscale representations than both a stan- |
dard MAE and a recently proposed, state-of-the-art MAEs |
SatMAE [13] and ConvMAE [21] across remote sensing |
datasets with a variety of scale and resolution characteristics. |
To the best of our knowledge Scale-MAE is the first self-supervised MAE to include scale-aware positional encoding |
and Laplacian pyramids. In our experiments, Scale-MAE |
achieves an average of a 5.6%nonparametric kNN classifica- |
tion improvement across eight remote sensing datasets com- |
pared to current state-of-the-art in addition to a 0.9mIoU |
to1.7mIoU improvement on the SpaceNet building seg- |
mentation transfer task for a range of evaluation scales (see |
Figure 1). |
2. Related Work |
Representation learning and the Masked Autoencoder. |
Representation learning aims to extract meaningful, intrin- |
sic features from data for downstream use [5]. In prac- |
tice, this often entails pretraining a deep network so that |
a lightweight learning routine can then finetune it for a par- |
ticular downstream task, see [15,16,17,24,27,30,37,49,66]. |
The Masked Autoencoder (MAE) is a recent state-of-the-art |
self-supervised representation learning method in computer |
vision that pretrains a ViT encoder by masking an image, |
feeding the unmasked portion into a transformer-based en- |
coder, and then tasking the decoder with reconstructing the |
input image [26]. MAEs fail to leverage scale information in |
scale-dependent domains as they are often reliant on absolute |
or relative positional encodings. To the best of our knowl- |
edge, Scale-MAE is the first MAE-based self-supervised |
learning method to incorporate a scale-variant positional |
encoding. |
Remote Sensing Representation Learning Neumann et |
al. [46] were one of the first to exhaustively share results on |
existing representation learning and semi-supervised learn- |
ing techniques for remote sensing imagery. Gao et al. [22] |
demonstrated the effectiveness of MAE pretraining for re- |
mote sensing image classification. Ayush et al. [3] lever- |
aged the metadata from remote sensing images via spatially |
aligned but temporally separated images as positive pairs |
for contrastive learning and predicted the latitude and longi- |
tude as pretext tasks. Gupta et al. [25] demonstrated the use |