Adobe Research developed the 3D Repoussé shape engine in Photoshop CS5 which allows smooth organic forms to be realized with ease from two dimensional curves. The system can be used by novice users to inflate font curves and hand drawn shapes to produce beautiful 3D design graphics. The surface is created by interactively solving a linearized thin plate spline model with carefully crafted boundary conditions providing a high level of user control.
Content-Aware Fill was introduced in Photoshop CS5. It dramatically simplifies the process of removing objects from an image. It can even be used to expand the boundaries of a cropped image, as shown in the before/after image to the left.
In this project, we consider the problem of video stabilization. We develop several technologies that are able to transform a video from a hand-held video camera so that it appears as if it were taken with a high-quality smooth camera motion. Different from traditional 2D stabilization methods, our technologies can adjust the video to appear as if it were taken from nearby viewpoints, allowing for 3D camera movements to be simulated. By aiming only for perceptual plausibility, rather than accurate reconstruction, we are able to develop algorithms that can effectively recreate dynamic scenes from a single source video. The Warp Stabilizer in After Effects (CS5.5 and later) is based on this research. View video >
Tutorial Builder is a Photoshop plug-in that automatically generates step-by-step Photoshop tutorials from a sequence of actions recorded from the application. The resulting tutorials include text descriptions, tool icons, and all the code required to replay each step in Photoshop using the Photoshop Connection SDK. Tutorial Builder was released on Adobe Labs in October 2011, and we are now releasing a Tutorial Player iPad app that displays and plays back Tutorial Builder content.
Our overarching research objective is to transform customer analytical needs into advanced analytical algorithms and services that generate high-value “advice” more autonomously and then push key findings to customers to enable them to automatically act upon those discoveries (action-based analytics). Our areas of research include:
Adobe Research is currently offering internships. We are looking for PhD students who are excited about pushing the state of the art in audio, graphics, vision, human-computer interaction, machine learning, visualization, analytics, optimization and more. We have just started recruiting, and we would love to hear from you!
Our researchers collaborate with top faculty and PhD students from universities around the world. By proactively engaging with thought leaders, Adobe contributes to and benefits from advances in the broader research community.