The authors model cropping preference as a field over possible crop boxes.
Continuous preference fields improve aesthetic crops beyond rigid annotation grids
The authors reconstruct dense aesthetic preferences from sparse crop labels, then use them to train and evaluate a vision-language cropping model.
Chinese Tech
Ziqing Zhang · Xiao Liu · Kai Liu · Jianze Li · Weihang Zhang · Linghe Kong · +1 more
Shanghai Jiao Tong University · Institute of Media Technology and Experience Design · Huawei
Research Digest··3 min read
Zhang et al.
Why this paper
From Huawei and 2 others
In one line
A continuous preference field for aesthetic cropping yields more accurate and generalizable crops than discrete grid annotation.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§