Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Searching for Efficient Multi-Scale Architectures for Dense Image Prediction

About

The design of neural network architectures is an important component for achieving state-of-the-art performance with machine learning systems across a broad array of tasks. Much work has endeavored to design and build architectures automatically through clever construction of a search space paired with simple learning algorithms. Recent progress has demonstrated that such meta-learning methods may exceed scalable human-invented architectures on image classification tasks. An open question is the degree to which such methods may generalize to new domains. In this work we explore the construction of meta-learning techniques for dense image prediction focused on the tasks of scene parsing, person-part segmentation, and semantic image segmentation. Constructing viable search spaces in this domain is challenging because of the multi-scale representation of visual information and the necessity to operate on high resolution imagery. Based on a survey of techniques in dense image prediction, we construct a recursive search space and demonstrate that even with efficient random search, we can identify architectures that outperform human-invented architectures and achieve state-of-the-art performance on three dense prediction tasks including 82.7\% on Cityscapes (street scene parsing), 71.3\% on PASCAL-Person-Part (person-part segmentation), and 87.9\% on PASCAL VOC 2012 (semantic image segmentation). Additionally, the resulting architecture is more computationally efficient, requiring half the parameters and half the computational cost as previous state of the art systems.

Liang-Chieh Chen, Maxwell D. Collins, Yukun Zhu, George Papandreou, Barret Zoph, Florian Schroff, Hartwig Adam, Jonathon Shlens• 2018

Related benchmarks

TaskDatasetResultRank
Semantic segmentationCityscapes (test)
mIoU82.7
1145
Semantic segmentationCityscapes (val)
mIoU80.8
332
Semantic segmentationCityscapes (val)
mIoU80.8
287
Human Part ParsingPASCAL-Person-Part (test)
mIoU71.34
68
Semantic segmentationCityscapes w/o coarse
mIoU82.7
29
Human ParsingPASCAL-Person-Part VOC 2010 (val)
mIoU71.34
13
Semantic segmentationCityscapes fine+coarse (test)
mIoU82.6
12
Semantic segmentationCityscapes 19 semantic class labels (val)
mIoU80.85
12
Semantic segmentationCityscapes (val)
mIoU80.9
11
Semantic Image SegmentationCityscapes
mIoU (%)82.7
8
Showing 10 of 10 rows

Other info

Code

Follow for update