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

Large-scale language-vision pre-training models, such as CLIP, have achievedremarkable text-guided image morphing results by leveraging severalunconditional generative models. However, existing CLIP-guided image morphingmethods encounter difficulties when morphing photorealistic images.Specifically, existing guidance fails to provide detailed explanations of themorphing regions within the image, leading to misguidance. In this paper, weobserved that such misguidance could be effectively mitigated by simply using aproper regularization loss. Our approach comprises two key components: 1) ageodesic cosine similarity loss that minimizes inter-modality features (i.e.,image and text) on a projected subspace of CLIP space, and 2) a latentregularization loss that minimizes intra-modality features (i.e., image andimage) on the image manifold. By replacing the na\"ive directional CLIP loss ina drop-in replacement manner, our method achieves superior morphing results onboth images and videos for various benchmarks, including CLIP-inversion.

On mitigating stability-plasticity dilemma in CLIP-guided image morphing via geodesic distillation loss