TY - GEN
T1 - Disregarding the big picture
T2 - 10th International Conference on Quality of Multimedia Experience, QoMEX 2018
AU - Wiedemann, Oliver
AU - Hosu, Vlad
AU - Lin, Hanhe
AU - Saupe, Dietmar
N1 - Funding Information:
We thank the German Research Foundation (DFG) for financial support within project A05 of SFB/Transregio 161.
Publisher Copyright:
© 2018 IEEE.
PY - 2018/9/11
Y1 - 2018/9/11
N2 - Image quality has been studied almost exclusively as a global image property. It is common practice for IQA databases and metrics to quantify this abstract concept with a single number per image. We propose an approach to blind IQA based on a convolutional neural network (patchnet) that was trained on a novel set of 32,000 individually annotated patches of 64×64 pixel. We use this model to generate spatially small local quality maps of images taken from KonIQ-10k, a large and diverse in-the-wild database of authentically distorted images. We show that our local quality indicator correlates well with global MOS, going beyond the predictive ability of quality related attributes such as sharpness. Averaging of patchnet predictions already outperforms classical approaches to global MOS prediction that were trained to include global image features. We additionally experiment with a generic second-stage aggregation CNN to estimate mean opinion scores. Our latter model performs comparable to the state of the art with a PLCC of 0.81 on KonIQ-10k.
AB - Image quality has been studied almost exclusively as a global image property. It is common practice for IQA databases and metrics to quantify this abstract concept with a single number per image. We propose an approach to blind IQA based on a convolutional neural network (patchnet) that was trained on a novel set of 32,000 individually annotated patches of 64×64 pixel. We use this model to generate spatially small local quality maps of images taken from KonIQ-10k, a large and diverse in-the-wild database of authentically distorted images. We show that our local quality indicator correlates well with global MOS, going beyond the predictive ability of quality related attributes such as sharpness. Averaging of patchnet predictions already outperforms classical approaches to global MOS prediction that were trained to include global image features. We additionally experiment with a generic second-stage aggregation CNN to estimate mean opinion scores. Our latter model performs comparable to the state of the art with a PLCC of 0.81 on KonIQ-10k.
UR - http://www.scopus.com/inward/record.url?scp=85054415262&partnerID=8YFLogxK
U2 - 10.1109/QoMEX.2018.8463384
DO - 10.1109/QoMEX.2018.8463384
M3 - Conference contribution
AN - SCOPUS:85054415262
SN - 978-1-5386-2606-1
T3 - 2018 10th International Conference on Quality of Multimedia Experience, QoMEX 2018
BT - 2018 10th International Conference on Quality of Multimedia Experience (QoMEX 2018)
PB - IEEE
Y2 - 29 May 2018 through 1 June 2018
ER -