English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The success of pre-training approaches on a variety of downstream tasks hasrevitalized the field of computer vision. Image aesthetics assessment (IAA) isone of the ideal application scenarios for such methods due to subjective andexpensive labeling procedure. In this work, an unified and flexible two-phase\textbf{C}LIP-based \textbf{S}emi-supervised \textbf{K}nowledge\textbf{D}istillation paradigm is proposed, namely \textbf{\textit{CSKD}}.Specifically, we first integrate and leverage a multi-source unlabeled datasetto align rich features between a given visual encoder and an off-the-shelf CLIPimage encoder via feature alignment loss. Notably, the given visual encoder isnot limited by size or structure and, once well-trained, it can seamlesslyserve as a better visual aesthetic learner for both student and teacher. In thesecond phase, the unlabeled data is also utilized in semi-supervised IAAlearning to further boost student model performance when applied inlatency-sensitive production scenarios. By analyzing the attention distance andentropy before and after feature alignment, we notice an alleviation of featurecollapse issue, which in turn showcase the necessity of feature alignmentinstead of training directly based on CLIP image encoder. Extensive experimentsindicate the superiority of CSKD, which achieves state-of-the-art performanceon multiple widely used IAA benchmarks.

CLIP Brings Better Features to Visual Aesthetics Learners