Presentation

<
PartEdit: Fine-Grained Image Editing using Pre-Trained Diffusion Models
SessionDeep Image Editing
DescriptionWe present PartEdit, a novel diffusion-based system enabling precise, text-based edits of object parts without retraining or manual masks. Optimizing part-aware tokens generates localized non-binary attention maps to guide seamless edits. Our novel blending strategy delivers high-quality visual results and outperforms prior techniques in both synthetic and real-world scenarios.
Authors
Aleksandar Cvejic
King Abdullah University of Science and Technology (KAUST)
Abdelrahman Eldesokey
King Abdullah University of Science and Technology (KAUST)
Peter Wonka
King Abdullah University of Science and Technology (KAUST)
Event Type
Technical Paper
TimeWednesday, 13 August 202511:35am - 11:45am PDT
LocationWest Building, Rooms 118-120
ACM Digital Library Journal Papers' PDFs
Conference Papers' PDFs
Session Time & Location
Sunday, 10 August 20256:00pm - 8:45pm PDTWest Building, Ballroom AB
Wednesday, 13 August 202510:45am - 12:35pm PDTWest Building, Rooms 118-120
Recordings
Livestreamed
Recorded
Next PresentationNext PresentationInstance Segmentation of Scene Sketches Using Natural Image Priors
Similar Presentations
Cora: Correspondence-aware image editing using few step diffusion
IntrinsicEdit: Precise generative image manipulation in intrinsic space
Nested Attention: Semantic-aware Attention Values for Concept Personalization