typos

[SimonZeng7108]

Hello authors,

Thanks a lot for your effort on this survey paper. It saved me a lot of time.

I spotted two typos in your current arXiv version (up to 31/01/2025).

1. Page 4, D. Convolution Layer, 'These
   sucecesses are supported by studies that have shown that
   perceptual distances are well modeled by deep activations
   within convolutional networks [27], [80]–[82].'
   Do you mean successes?

2. Page 11, Fig 7, No reference, Low-level Features, the reference number for FasterVQA is wrong, it is supposed to be 263 instead of 267?

A quick question: why did you mention Patch-VQ as a blind VQA method but did not refer to them in UGC datasets? They are the largest UGC dataset so far?

best,

Simon