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/

Zendy Plus

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

Zendy Tools

Zendy Open

We present HawkI, for synthesizing aerial-view images from text and anexemplar image, without any additional multi-view or 3D information forfinetuning or at inference. HawkI uses techniques from classical computervision and information theory. It seamlessly blends the visual features fromthe input image within a pretrained text-to-2Dimage stable diffusion model witha test-time optimization process for a careful bias-variance trade-off, whichuses an Inverse Perspective Mapping (IPM) homography transformation to providesubtle cues for aerialview synthesis. At inference, HawkI employs a uniquemutual information guidance formulation to steer the generated image towardsfaithfully replicating the semantic details of the input-image, whilemaintaining a realistic aerial perspective. Mutual information guidancemaximizes the semantic consistency between the generated image and the inputimage, without enforcing pixel-level correspondence between vastly differentviewpoints. Through extensive qualitative and quantitative comparisons againsttext + exemplar-image based methods and 3D/ multi-view based novel-viewsynthesis methods on proposed synthetic and real datasets, we demonstrate thatour method achieves a significantly better bias-variance trade-off towardsgenerating high fidelity aerial-view images.Code and data is available athttps://github.com/divyakraman/HawkI2024.

HawkI: Homography & Mutual Information Guidance for 3D-free Single Image to Aerial View