Transferring Image-based Edits for Multi-Channel Compositing

1University College London     2Adobe Research

Conditionally Accepted: SIGGRAPH-Asia 2017

(Top) Input source view rendered using a set of photometric render channels. (a) Composite of All Photometric channels. (b)The user applies 2D image-based edits to specified channels such as: blurring the background object to create depth of field effect (All Photometric channels); adjusting gamma, hue, and saturation to emphasise floor reflections (Reflection channel); Making the eye sockets of foreground skulls appear to glow blue by adjusting the hue, saturation, and lightness (Diffuse and Global Illum channels). (Bottom) Given a target view (a) with a different scene configuration (skulls are positioned in different 3D locations and orientations) (b) our method transfers the 2D image-based user edits automatically. The right column (c) shows the outlines of the corresponding localization masks for the two views. Multiple instances of the same object make this a challenging scene.


A common way to generate high-quality product images is to start with a physically-based render of a 3D scene, apply image-based edits on individual render channels, and then composite the edited channels together (in some cases, on top of a background photograph). This workflow requires users to manually select the right render channels, prescribe channel-specific masks, and set appropriate edit parameters. Unfortunately, such edits cannot be easily reused for global variations of the original scene, such as a rigid-body transformation of the 3D objects or a modified viewpoint, which discourages iterative refinement of both global scene changes and image-based edits. We propose a method to automatically transfer such user edits across variations of object geometry, illumination, and viewpoint. This transfer problem is challenging since many edits may be visually plausible but non-physical, with a successful transfer dependent on an unknown set of scene attributes that may include both photometric and non-photometric features. The problem of transferring edits is challenging as on the one hand they often involve multiple channels, while on the other hand adding too many channels can easily result in corrupted transfers. To address this challenge, we present a transfer algorithm that extends the image analogies formulation to include an augmented set of photometric and non-photometric guidance channels and, more importantly, adaptively estimate weights for the various candidate channels in a way that matches the characteristics of each individual edit. We demonstrate our algorithm on a variety of complex edit-transfer scenarios for creating high-quality product images.