Yifan Wang, Brian Curless, Steve Seitz
By analyzing the motion of people and other objects in a scene, we demonstrate how to infer depth, occlusion, lighting, and shadow information from video taken from a single camera viewpoint. This information is then used to composite new objects into the same scene with a high degree of automation and realism. In particular, when a user places a new object (2D cut-out) in the image, it is automatically rescaled, relit, occluded properly, and casts realistic shadows in the correct direction relative to the sun, and which conform properly to scene geometry. We demonstrate results (best viewed in supplementary video) on a range of scenes and compare to alternative methods for depth estimation and shadow compositing.
Ricardo Martin-Brualla, Rohit Pandey, Shuoran Yang, Pavel Pidlypenskyi, Jonathan Taylor, Julien Valentin, Sameh Khamis, Philip Davidson, Anastasia Tkach, Peter Lincoln, Adarsh Kowdle, Christoph Rhemann, Dan B Goldman, Cem Keskin, Steve Seitz, Shahram Izadi, Sean Fanello
Motivated by augmented and virtual reality applications such as telepresence, there has been a recent focus in real-time performance capture of humans under motion. However, given the real-time constraint, these systems often suffer from artifacts in geometry and texture such as holes and noise in the final rendering, poor lighting, and low-resolution textures. We take the novel approach to augment such real-time performance capture systems with a deep architecture that takes a rendering from an arbitrary viewpoint, and jointly performs completion, super resolution, and denoising of the imagery in real-time. We call this approach neural (re-)rendering, and our live system "LookinGood". Our deep architecture is trained to produce high resolution and high quality images from a coarse rendering in real-time. First, we propose a self-supervised training method that does not require manual ground-truth annotation. We contribute a specialized reconstruction error that uses semantic information to focus on relevant parts of the subject, e.g. the face. We also introduce a salient reweighing scheme of the loss function that is able to discard outliers. We specifically design the system for virtual and augmented reality headsets where the consistency between the left and right eye plays a crucial role in the final user experience. Finally, we generate temporally stable results by explicitly minimizing the difference between two consecutive frames. We tested the proposed system in two different scenarios: one involving a single RGB-D sensor, and upper body reconstruction of an actor, the second consisting of full body 360 degree capture. Through extensive experimentation, we demonstrate how our system generalizes across unseen sequences and subjects. The supplementary video is available at http://youtu.be/Md3tdAKoLGU.
Shanchuan Lin, Andrey Ryabtsev, Soumyadip Sengupta, Brian Curless, Steve Seitz, Ira Kemelmacher-Shlizerman
We introduce a real-time, high-resolution background replacement technique which operates at 30fps in 4K resolution, and 60fps for HD on a modern GPU. Our technique is based on background matting, where an additional frame of the background is captured and used in recovering the alpha matte and the foreground layer. The main challenge is to compute a high-quality alpha matte, preserving strand-level hair details, while processing high-resolution images in real-time. To achieve this goal, we employ two neural networks; a base network computes a low-resolution result which is refined by a second network operating at high-resolution on selective patches. We introduce two largescale video and image matting datasets: VideoMatte240K and PhotoMatte13K/85. Our approach yields higher quality results compared to the previous state-of-the-art in background matting, while simultaneously yielding a dramatic boost in both speed and resolution.
Jeong Joon Park, Aleksander Holynski, Steve Seitz
We address the dual problems of novel view synthesis and environment reconstruction from hand-held RGBD sensors. Our contributions include 1) modeling highly specular objects, 2) modeling inter-reflections and Fresnel effects, and 3) enabling surface light field reconstruction with the same input needed to reconstruct shape alone. In cases where scene surface has a strong mirror-like material component, we generate highly detailed environment images, revealing room composition, objects, people, buildings, and trees visible through windows. Our approach yields state of the art view synthesis techniques, operates on low dynamic range imagery, and is robust to geometric and calibration errors.
Supasorn Suwajanakorn, Ira Kemelmacher-Shlizerman, Steve Seitz
We reconstruct a controllable model of a person from a large photo collection that captures his or her {\em persona}, i.e., physical appearance and behavior. The ability to operate on unstructured photo collections enables modeling a huge number of people, including celebrities and other well photographed people without requiring them to be scanned. Moreover, we show the ability to drive or {\em puppeteer} the captured person B using any other video of a different person A. In this scenario, B acts out the role of person A, but retains his/her own personality and character. Our system is based on a novel combination of 3D face reconstruction, tracking, alignment, and multi-texture modeling, applied to the puppeteering problem. We demonstrate convincing results on a large variety of celebrities derived from Internet imagery and video.
Xiaojuan Wang, Janne Kontkanen, Brian Curless, Steve Seitz, Ira Kemelmacher, Ben Mildenhall, Pratul Srinivasan, Dor Verbin, Aleksander Holynski
We present a method that uses a text-to-image model to generate consistent content across multiple image scales, enabling extreme semantic zooms into a scene, e.g., ranging from a wide-angle landscape view of a forest to a macro shot of an insect sitting on one of the tree branches. We achieve this through a joint multi-scale diffusion sampling approach that encourages consistency across different scales while preserving the integrity of each individual sampling process. Since each generated scale is guided by a different text prompt, our method enables deeper levels of zoom than traditional super-resolution methods that may struggle to create new contextual structure at vastly different scales. We compare our method qualitatively with alternative techniques in image super-resolution and outpainting, and show that our method is most effective at generating consistent multi-scale content.
Mengyi Shan, Brian Curless, Ira Kemelmacher-Shlizerman, Steve Seitz
We challenge text-to-image models with generating escape room puzzle images that are visually appealing, logically solid, and intellectually stimulating. While base image models struggle with spatial relationships and affordance reasoning, we propose a hierarchical multi-agent framework that decomposes this task into structured stages: functional design, symbolic scene graph reasoning, layout synthesis, and local image editing. Specialized agents collaborate through iterative feedback to ensure the scene is visually coherent and functionally solvable. Experiments show that agent collaboration improves output quality in terms of solvability, shortcut avoidance, and affordance clarity, while maintaining visual quality.
Soumyadip Sengupta, Vivek Jayaram, Brian Curless, Steve Seitz, Ira Kemelmacher-Shlizerman
We propose a method for creating a matte -- the per-pixel foreground color and alpha -- of a person by taking photos or videos in an everyday setting with a handheld camera. Most existing matting methods require a green screen background or a manually created trimap to produce a good matte. Automatic, trimap-free methods are appearing, but are not of comparable quality. In our trimap free approach, we ask the user to take an additional photo of the background without the subject at the time of capture. This step requires a small amount of foresight but is far less time-consuming than creating a trimap. We train a deep network with an adversarial loss to predict the matte. We first train a matting network with supervised loss on ground truth data with synthetic composites. To bridge the domain gap to real imagery with no labeling, we train another matting network guided by the first network and by a discriminator that judges the quality of composites. We demonstrate results on a wide variety of photos and videos and show significant improvement over the state of the art.
Jeong Joon Park, Richard Newcombe, Steve Seitz
We present an approach for interactively scanning highly reflective objects with a commodity RGBD sensor. In addition to shape, our approach models the surface light field, encoding scene appearance from all directions. By factoring the surface light field into view-independent and wavelength-independent components, we arrive at a representation that can be robustly estimated with IR-equipped commodity depth sensors, and achieves high quality results.
Mengyi Shan, Brian Curless, Ira Kemelmacher-Shlizerman, Steve Seitz
We present a system that automatically brings street view imagery to life by populating it with naturally behaving, animated pedestrians and vehicles. Our approach is to remove existing people and vehicles from the input image, insert moving objects with proper scale, angle, motion, and appearance, plan paths and traffic behavior, as well as render the scene with plausible occlusion and shadowing effects. The system achieves these by reconstructing the still image street scene, simulating crowd behavior, and rendering with consistent lighting, visibility, occlusions, and shadows. We demonstrate results on a diverse range of street scenes including regular still images and panoramas.
Konstantinos Rematas, Ira Kemelmacher-Shlizerman, Brian Curless, Steve Seitz
We present a system that transforms a monocular video of a soccer game into a moving 3D reconstruction, in which the players and field can be rendered interactively with a 3D viewer or through an Augmented Reality device. At the heart of our paper is an approach to estimate the depth map of each player, using a CNN that is trained on 3D player data extracted from soccer video games. We compare with state of the art body pose and depth estimation techniques, and show results on both synthetic ground truth benchmarks, and real YouTube soccer footage.
Xuan Luo, Yanmeng Kong, Jason Lawrence, Ricardo Martin-Brualla, Steve Seitz
This paper introduces the largest and most diverse collection of rectified stereo image pairs to the research community, KeystoneDepth, consisting of tens of thousands of stereographs of historical people, events, objects, and scenes between 1860 and 1963. Leveraging the Keystone-Mast raw scans from the California Museum of Photography, we apply multiple processing steps to produce clean stereo image pairs, complete with calibration data, rectification transforms, and depthmaps. A second contribution is a novel approach for view synthesis that runs at real-time rates on a mobile device, simulating the experience of looking through an open window into these historical scenes. We produce results for thousands of antique stereographs, capturing many important historical moments.
Teerapat Jenrungrot, Vivek Jayaram, Steve Seitz, Ira Kemelmacher-Shlizerman
Given a multi-microphone recording of an unknown number of speakers talking concurrently, we simultaneously localize the sources and separate the individual speakers. At the core of our method is a deep network, in the waveform domain, which isolates sources within an angular region $θ\pm w/2$, given an angle of interest $θ$ and angular window size $w$. By exponentially decreasing $w$, we can perform a binary search to localize and separate all sources in logarithmic time. Our algorithm allows for an arbitrary number of potentially moving speakers at test time, including more speakers than seen during training. Experiments demonstrate state-of-the-art performance for both source separation and source localization, particularly in high levels of background noise.
Soumyadip Sengupta, Brian Curless, Ira Kemelmacher-Shlizerman, Steve Seitz
Every time you sit in front of a TV or monitor, your face is actively illuminated by time-varying patterns of light. This paper proposes to use this time-varying illumination for synthetic relighting of your face with any new illumination condition. In doing so, we take inspiration from the light stage work of Debevec et al., who first demonstrated the ability to relight people captured in a controlled lighting environment. Whereas existing light stages require expensive, room-scale spherical capture gantries and exist in only a few labs in the world, we demonstrate how to acquire useful data from a normal TV or desktop monitor. Instead of subjecting the user to uncomfortable rapidly flashing light patterns, we operate on images of the user watching a YouTube video or other standard content. We train a deep network on images plus monitor patterns of a given user and learn to predict images of that user under any target illumination (monitor pattern). Experimental evaluation shows that our method produces realistic relighting results. Video results are available at http://grail.cs.washington.edu/projects/Light_Stage_on_Every_Desk/.
Luyang Zhu, Konstantinos Rematas, Brian Curless, Steve Seitz, Ira Kemelmacher-Shlizerman
Great progress has been made in 3D body pose and shape estimation from a single photo. Yet, state-of-the-art results still suffer from errors due to challenging body poses, modeling clothing, and self occlusions. The domain of basketball games is particularly challenging, as it exhibits all of these challenges. In this paper, we introduce a new approach for reconstruction of basketball players that outperforms the state-of-the-art. Key to our approach is a new method for creating poseable, skinned models of NBA players, and a large database of meshes (derived from the NBA2K19 video game), that we are releasing to the research community. Based on these models, we introduce a new method that takes as input a single photo of a clothed player in any basketball pose and outputs a high resolution mesh and 3D pose for that player. We demonstrate substantial improvement over state-of-the-art, single-image methods for body shape reconstruction.
Ira Kemelmacher-Shlizerman, Steve Seitz, Daniel Miller, Evan Brossard
Recent face recognition experiments on a major benchmark LFW show stunning performance--a number of algorithms achieve near to perfect score, surpassing human recognition rates. In this paper, we advocate evaluations at the million scale (LFW includes only 13K photos of 5K people). To this end, we have assembled the MegaFace dataset and created the first MegaFace challenge. Our dataset includes One Million photos that capture more than 690K different individuals. The challenge evaluates performance of algorithms with increasing numbers of distractors (going from 10 to 1M) in the gallery set. We present both identification and verification performance, evaluate performance with respect to pose and a person's age, and compare as a function of training data size (number of photos and people). We report results of state of the art and baseline algorithms. Our key observations are that testing at the million scale reveals big performance differences (of algorithms that perform similarly well on smaller scale) and that age invariant recognition as well as pose are still challenging for most. The MegaFace dataset, baseline code, and evaluation scripts, are all publicly released for further experimentations at: megaface.cs.washington.edu.
Xiaojuan Wang, Aleksander Holynski, Brian Curless, Ira Kemelmacher, Steve Seitz
We present a framework for generating music-synchronized, choreography aware animal dance videos. Our framework introduces choreography patterns -- structured sequences of motion beats that define the long-range structure of a dance -- as a novel high-level control signal for dance video generation. These patterns can be automatically estimated from human dance videos. Starting from a few keyframes representing distinct animal poses, generated via text-to-image prompting or GPT-4o, we formulate dance synthesis as a graph optimization problem that seeks the optimal keyframe structure to satisfy a specified choreography pattern of beats. We also introduce an approach for mirrored pose image generation, essential for capturing symmetry in dance. In-between frames are synthesized using an video diffusion model. With as few as six input keyframes, our method can produce up to 30 seconds dance videos across a wide range of animals and music tracks.
Susung Hong, Brian Curless, Ira Kemelmacher-Shlizerman, Steve Seitz
We propose a fully automated AI system that produces short comedic videos similar to sketch shows such as Saturday Night Live. Starting with character references, the system employs a population of agents loosely based on real production studio roles, structured to optimize the quality and diversity of ideas and outputs through iterative competition, evaluation, and improvement. A key contribution is the introduction of LLM critics aligned with real viewer preferences through the analysis of a corpus of comedy videos on YouTube to automatically evaluate humor. Our experiments show that our framework produces results approaching the quality of professionally produced sketches while demonstrating state-of-the-art performance in video generation.
Chris Sweeney, Aleksander Holynski, Brian Curless, Steve M Seitz
We present a novel Structure from Motion pipeline that is capable of reconstructing accurate camera poses for panorama-style video capture without prior camera intrinsic calibration. While panorama-style capture is common and convenient, previous reconstruction methods fail to obtain accurate reconstructions due to the rotation-dominant motion and small baseline between views. Our method is built on the assumption that the camera motion approximately corresponds to motion on a sphere, and we introduce three novel relative pose methods to estimate the fundamental matrix and camera distortion for spherical motion. These solvers are efficient and robust, and provide an excellent initialization for bundle adjustment. A soft prior on the camera poses is used to discourage large deviations from the spherical motion assumption when performing bundle adjustment, which allows cameras to remain properly constrained for optimization in the absence of well-triangulated 3D points. To validate the effectiveness of the proposed method we evaluate our approach on both synthetic and real-world data, and demonstrate that camera poses are accurate enough for multiview stereo.