Image Super-Resolution

The cluster focuses on discussions about super-resolution and AI-based image upscaling techniques, comparing their performance to traditional methods like bicubic interpolation, and addressing issues like information loss or hallucinations in results.

📉 Falling 0.5x AI & Machine Learning
2,887
Comments
20
Years Active
5
Top Authors
#6972
Topic ID

Activity Over Time

2007
2
2008
2
2009
4
2010
35
2011
37
2012
86
2013
77
2014
72
2015
108
2016
139
2017
157
2018
95
2019
138
2020
308
2021
340
2022
274
2023
394
2024
311
2025
290
2026
18

Keywords

e.g DLSS AI NN HN ML MFG ST CSI SR resolution image images interpolation super pixel blurry enhance hallucination information

Sample Comments

amelius May 4, 2021 View on HN

Was hoping they used super-resolution magic to scale up the image.

msoad Jul 21, 2022 View on HN

Photoshop also has super resolution

p1necone Feb 5, 2020 View on HN

The upscaled version looks worse to me. Seems like it's lost information, not gained.

fhk Aug 17, 2023 View on HN

Have you thought of running super resolution on the images?

SubiculumCode Jun 14, 2019 View on HN

Almost like super resolution techniques?

WithinReason Mar 23, 2023 View on HN

What's wrong with upscaling?

gnopgnip Oct 14, 2021 View on HN

You can't really use machine learning based image upscaling for that type of work

Ph0X Oct 13, 2016 View on HN

Isn't this an issue with the algorithm?Can someone try and see how it would perform if you simply upscale the image using normal bicubic interpolation? And if it performs much better, I feel like that should be a preprocessing option to scale up the image since it seems to do so poorly on small resolutions.

mgraczyk Dec 29, 2025 View on HN

You're mistaken and the original experiment does not distinguish between classic edge aware upscaling/super resolution vs more problematic replacement

basch Mar 21, 2019 View on HN

Its not upscaling. Its taking what it knows about other non ST pictures and creating new texture and information.