Is image super-resolution helpful for other vision tasks?

Dai, Dengxin; Wang, Yujian; Chen, Yuhua; Gool, Luc Van

doi:10.1109/wacv.2016.7477613

Cited by 118 publications

(67 citation statements)

References 44 publications

(93 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Image scale has shown to play a vital role in many computer vision tasks [3,2,6]. To further analyze the impact of image-level and instance-level adaptation, we conduct experiment on KITTI → Cityscapes by varying the image scales.…”

Section: Image-level Vs Instance-level Alignmentmentioning

confidence: 99%

Domain Adaptive Faster R-CNN for Object Detection in the Wild

Chen

Wen

Sakaridis

et al. 2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

Self Cite

1,334

1,322

View full text Add to dashboard Cite

Object detection typically assumes that training and test data are drawn from an identical distribution, which, however, does not always hold in practice. Such a distribution mismatch will lead to a significant performance drop. In this work, we aim to improve the cross-domain robustness of object detection. We tackle the domain shift on two levels: 1) the image-level shift, such as image style, illumination, etc., and 2) the instance-level shift, such as object appearance, size, etc. We build our approach based on the recent state-of-the-art Faster R-CNN model, and design two domain adaptation components, on image level and instance level, to reduce the domain discrepancy. The two domain adaptation components are based on H-divergence theory, and are implemented by learning a domain classifier in adversarial training manner. The domain classifiers on different levels are further reinforced with a consistency regularization to learn a domain-invariant region proposal network (RPN) in the Faster R-CNN model. We evaluate our newly proposed approach using multiple datasets including Cityscapes, KITTI, SIM10K, etc. The results demonstrate the effectiveness of our proposed approach for robust object detection in various domain shift scenarios.

show abstract

Section: Image-level Vs Instance-level Alignmentmentioning

confidence: 99%

Domain Adaptive Faster R-CNN for Object Detection in the Wild

Chen

Wen

Sakaridis

et al. 2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

Self Cite

1,334

1,322

View full text Add to dashboard Cite

show abstract

“…During training, the CNN features extracted from the original images are treated as privileged information, and the CNN features extracted from the downsampled images as the main features. The experimental setting is inspired by [6], where high-resolution images yield superior performance than low-resolution images for standard vision tasks. PCA is applied on two types of features to obtain 100-d compact representations.…”

Section: Image Classificationmentioning

confidence: 99%

Fast Algorithms for Linear and Kernel SVM+

Dai

Tan

et al. 2016

2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

View full text Add to dashboard Cite

“…Although several studies have been conducted using SR as a pre-processing step [1,11,12,33,42,3,10,5], none have quantified its affect on object detection performance in satellite imagery across multiple resolutions. This study aims to accomplish that task by training multiple custom object detection models to identify vehicles, boats, and planes in both native and super-resolved data.…”

Section: Introductionmentioning

confidence: 99%

The Effects of Super-Resolution on Object Detection Performance in Satellite Imagery

Shermeyer

Etten

2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

190

View full text Add to dashboard Cite

We explore the application of super-resolution techniques to satellite imagery, and the effects of these techniques on object detection algorithm performance. Specifically, we enhance satellite imagery beyond its native resolution, and test if we can identify various types of vehicles, planes, and boats with greater accuracy than native resolution. Using the Very Deep Super-Resolution (VDSR) framework and a custom Random Forest Super-Resolution (RFSR) framework we generate enhancement levels of 2×, 4×, and 8× over five distinct resolutions ranging from 30 cm to 4.8 meters. Using both native and super-resolved data, we then train several custom detection models using the SIMRDWN object detection framework. SIMRDWN combines a number of popular object detection algorithms (e.g. SSD, YOLO) into a unified framework that is designed to rapidly detect objects in large satellite images. This approach allows us to quantify the effects of super-resolution techniques on object detection performance across multiple classes and resolutions. We also quantify the performance of object detection as a function of native resolution and object pixel size. For our test set we note that performance degrades from mean average precision (mAP) = 0.53 at 30 cm resolution, down to mAP = 0.11 at 4.8 m resolution. Super-resolving native 30 cm imagery to 15 cm yields the greatest benefit; a 13 − 36% improvement in mAP. Superresolution is less beneficial at coarser resolutions, though still provides a small improvement in performance.

show abstract

Is image super-resolution helpful for other vision tasks?

Cited by 118 publications

References 44 publications

Domain Adaptive Faster R-CNN for Object Detection in the Wild

Domain Adaptive Faster R-CNN for Object Detection in the Wild

Fast Algorithms for Linear and Kernel SVM+

The Effects of Super-Resolution on Object Detection Performance in Satellite Imagery

Contact Info

Product

Resources

About