Beyond Realism: Learning the Art of Expressive Composition with StickerNet
Haoming Lu, David Kocharian, Humphrey Shi
TL;DR
This work introduces expressive composition, a non-photorealistic image composition paradigm rooted in real-world user edits that emphasize style and intent over realism. It builds a large-scale dataset of 1.8 million community-driven edits and proposes StickerNet, a lightweight two-stage model that first classifies composition type and then predicts type-specific placement, enabling efficient on-device operation. Empirical results from user studies and quantitative analyses show StickerNet closely matches human placement behavior and outperforms standard baselines, validating learning from authentic editing actions. The approach broadens visual understanding toward expressive, user-driven creativity with practical implications for creative platforms and edge devices.
Abstract
As a widely used operation in image editing workflows, image composition has traditionally been studied with a focus on achieving visual realism and semantic plausibility. However, in practical editing scenarios of the modern content creation landscape, many compositions are not intended to preserve realism. Instead, users of online platforms motivated by gaining community recognition often aim to create content that is more artistic, playful, or socially engaging. Taking inspiration from this observation, we define the expressive composition task, a new formulation of image composition that embraces stylistic diversity and looser placement logic, reflecting how users edit images on real-world creative platforms. To address this underexplored problem, we present StickerNet, a two-stage framework that first determines the composition type, then predicts placement parameters such as opacity, mask, location, and scale accordingly. Unlike prior work that constructs datasets by simulating object placements on real images, we directly build our dataset from 1.8 million editing actions collected on an anonymous online visual creation and editing platform, each reflecting user-community validated placement decisions. This grounding in authentic editing behavior ensures strong alignment between task definition and training supervision. User studies and quantitative evaluations show that StickerNet outperforms common baselines and closely matches human placement behavior, demonstrating the effectiveness of learning from real-world editing patterns despite the inherent ambiguity of the task. This work introduces a new direction in visual understanding that emphasizes expressiveness and user intent over realism.