Jump to Content
William T. Freeman

William T. Freeman

Bill Freeman is a Senior Research Scientist at Google, managing a team within Machine Perception doing research in vision and graphics. He is also a faculty member at MIT, in the Electrical Engineering and Computer Science Department, and a member of CSAIL, the Computer Science and Artificial intelligence Laboratory there. He received outstanding paper awards at computer vision or machine learning conferences in 1997, 2006, 2009 and 2012, and test-of-time awards for papers from 1990 and 1995.
Authored Publications
Google Publications
Other Publications
Sort By
  • Title
  • Title, descending
  • Year
  • Year, descending
    Preview abstract Creativity is an indispensable part of human cognition and also an inherent part of how we make sense of the world. Metaphorical abstraction is fundamental in communicating creative ideas through nuanced relationships between abstract concepts such as feelings. While computer vision benchmarks and approaches predominantly focus on understanding and generating literal interpretations of images, metaphorical comprehension of images remains relatively unexplored. Towards this goal, we introduce MetaCLUE, a set of vision tasks on visual metaphor. We also collect high-quality and rich metaphor annotations (abstract objects, concepts, relationships along with their corresponding object boxes) as there do not exist any datasets that facilitate the evaluation of these tasks. We perform a comprehensive analysis of state-of-the-art models in vision and language based on our annotations, highlighting strengths and weaknesses of current approaches in visual metaphor Classification, Localization, Understanding (retrieval, question answering, captioning) and gEneration (text-to-image synthesis) tasks. We hope this work provides a concrete step towards developing AI systems with human-like creative capabilities. View details
    MaskGIT: Masked Image Generative Transformers
    Huiwen Chang
    Han Zhang
    Lu Jiang
    Ce Liu
    CVPR (2022)
    Preview abstract Generative transformers have experienced rapid popularity growth in the computer vision community in synthesizing high-fidelity and high-resolution images. The best generative transformer models so far, however, still treat an image naively as a sequence of tokens, and decode an image sequentially following the raster scan ordering (i.e. line-by-line). We find this strategy neither optimal nor efficient. This paper proposes a novel image synthesis paradigm using a bidirectional transformer decoder, which we term MaskGIT. During training, MaskGIT learns to predict randomly masked tokens by attending to tokens in all directions. At inference time, the model begins with generating all tokens of an image simultaneously, and then refines the image iteratively conditioned on the previous generation. Our experiments demonstrate that MaskGIT significantly outperforms the state-of-the-art transformer model on the ImageNet dataset, and accelerates autoregressive decoding by up to 64x. Besides, we illustrate that MaskGIT can be easily extended to various image editing tasks, such as inpainting, extrapolation, and image manipulation. View details
    Preview abstract Single image 3D photography enables viewers to view a still image from novel viewpoints. Recent approaches for single-image view synthesis combine monocular depth network along with inpainting networks resulting in compelling novel view synthesis results. A drawback of these approaches is the use of hard layering making them not suitable to model intricate appearance effects such as matting. We present SLIDE, a modular and unified system for single image 3D photography that uses simple yet effective soft layering strategy to model appearance effects. In addition, we propose a novel depth-aware training of inpainting network suitable for 3D photography task. Extensive experimental analysis on 3 different view synthesis datasets in combination with user studies on in-the-wild image collections demonstrate the superior performance of our technique in comparison to existing strong baselines. View details
    THUNDR: Transformer-based 3D HUmaN Reconstruction with Markers
    Mihai Zanfir
    Andrei Zanfir
    Proceedings of the IEEE/CVF International Conference on Computer Vision (2021)
    Preview abstract We present THUNDR, a transformer-based deep neural network methodology to reconstruct the 3D pose and shape of people, given monocular RGB images. Key to our methodology is an intermediate 3D marker representation, where we aim to combine the predictive power of model-free output architectures and the regularizing, anthropometrically-preserving properties of a statistical human surface models like GHUM—a recently introduced, expressive full body statistical 3d human model, trained end-to-end. Our novel transformer-based prediction pipeline can focus on image regions relevant to the task, supports self-supervised regimes, and ensures that solutions are consistent with human anthropometry. We show state-of-the-art results on Human3.6M and 3DPW, for both the fully-supervised and the self-supervised models, for the task of inferring 3D human shape, joint positions, and global translation. Moreover, we observe very solid 3d reconstruction performance for difficult human poses collected in the wild. Models will be made available for research. View details
    AutoFlow: Learning a Better Training Set for Optical Flow
    Daniel Vlasic
    Charles Herrmann
    Varun Jampani
    Huiwen Chang
    Ramin Zabih
    Ce Liu
    (2021)
    Preview abstract Synthetic datasets play a critical role in pre-training CNN models for optical flow, but they are painstaking to generate and hard to adapt to new applications. To automate the process, we present AutoFlow, a simple and effective method to render training data for optical flow that optimizes the performance of a model on a target dataset. AutoFlow takes a layered approach to render synthetic data, where the motion, shape, and appearance of each layer are controlled by learnable hyperparameters. Experimental results show that AutoFlow achieves state-of-the-art accuracy in pre-training both PWC-Net and RAFT. Our code and data are available at https://autoflow-google.github.io. View details
    Neural Descent for Visual 3D Human Pose and Shape
    Andrei Zanfir
    Mihai Zanfir
    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2021), pp. 14484-14493
    Preview abstract We present a deep neural network methodology to reconstruct the 3d pose and shape of people, given image or video inputs. We rely on a recently introduced, expressive full body statistical 3d human model, GHUM, with facial expression and hand detail and aim to learn to reconstruct the model pose and shape states in a self-supervised regime. Central to our methodology, is a learning to learn approach, referred to as HUman Neural Descent (HUND) that avoids both second-order differentiation when training the model parameters, and expensive state gradient descent in order to accurately minimize a semantic differentiable rendering loss at test time. Instead, we rely on novel recurrent stages to update the pose and shape parameters such that not only losses are minimized effectively but the process is regularized in order to ensure progress. The newly introduced architecture is tested extensively, and achieves state-of-the-art results on datasets like H3.6M and 3DPW, as well as in complex imagery collected in-the-wild. View details
    Neural Light Transport for Relighting and View Synthesis
    Xiuming Zhang
    Yun-Ta Tsai
    Tiancheng Sun
    Tianfan Xue
    Philip Davidson
    Christoph Rhemann
    Paul Debevec
    Ravi Ramamoorthi
    ACM Transactions on Graphics, vol. 40 (2021)
    Preview abstract The light transport (LT) of a scene describes how it appears under different lighting and viewing directions, and complete knowledge of a scene's LT enables the synthesis of novel views under arbitrary lighting. In this paper, we focus on image-based LT acquisition, primarily for human bodies within a light stage setup. We propose a semi-parametric approach to learn a neural representation of LT that is embedded in the space of a texture atlas of known geometric properties, and model all non-diffuse and global LT as residuals added to a physically-accurate diffuse base rendering. In particular, we show how to fuse previously seen observations of illuminants and views to synthesize a new image of the same scene under a desired lighting condition from a chosen viewpoint. This strategy allows the network to learn complex material effects (such as subsurface scattering) and global illumination, while guaranteeing the physical correctness of the diffuse LT (such as hard shadows). With this learned LT, one can relight the scene photorealistically with a directional light or an HDRI map, synthesize novel views with view-dependent effects, or do both simultaneously, all in a unified framework using a set of sparse, previously seen observations. Qualitative and quantitative experiments demonstrate that our neural LT (NLT) outperforms state-of-the-art solutions for relighting and view synthesis, without separate treatment for both problems that prior work requires. View details
    Preview abstract Image classification models can depend on multiple different semantic attributes of the image. An explanation of the decision of the classifier needs to both discover and visualize these properties. Here we present StylEx, a method for doing this, by training a generative model to specifically explain multiple attributes that underlie classifier decisions. A natural source for such attributes is the S-space of StyleGAN, which is known to generate semantically meaningful dimensions in the image. However, these will typically not correspond to classifier-specific attributes since standard GAN training is not dependent on the classifier. To overcome this, we propose training procedure for a StyleGAN, which incorporates the classifier model. This results in an S-space that captures distinct attributes underlying classifier outputs. After training, the model can be used to visualize the effect of changing multiple attributes per image, thus providing an image-specific explanation. We apply StylEx to multiple domains, including animals, leaves, faces and retinal images. For these, we show how an image can be changed in different ways to change its classifier prediction. Our results show that the method finds attributes that align well with semantic ones, generate meaningful image-specific explanations, and are interpretable as measured in user-studies. View details
    GHUM & GHUML: Generative 3D Human Shape and Articulated Pose Models
    Hongyi Xu
    Andrei Zanfir
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (Oral) (2020), pp. 6184-6193
    Preview abstract We present a statistical, articulated 3D human shape modeling pipeline, within a fully trainable, modular, deep learning framework. Given high-resolution complete 3D body scans of humans, captured in various poses, together with additional closeups of their head and facial expressions, as well as hand articulation, and given initial, artist designed, gender neutral rigged quad-meshes, we train all model parameters including non-linear shape spaces based on variational auto-encoders, pose-space deformation correctives, skeleton joint center predictors, and blend skinning functions, in a single consistent learning loop. The models are simultaneously trained with all the 3d dynamic scan data (over60,000diverse human configurations in our new dataset) in order to capture correlations and en-sure consistency of various components. Models support facial expression analysis, as well as body (with detailed hand) shape and pose estimation. We provide fully train-able generic human models of different resolutions – the moderate-resolution GHUM consisting of 10,168 vertices and the low-resolution GHUML(ite) of 3,194 vertices –, run comparisons between them, analyze the impact of different components and illustrate their reconstruction from image data. The models are available for research. View details
    Preview abstract Monocular 3D human pose and shape estimation is challenging due to the many degrees of freedom of the human body and thedifficulty to acquire training data for large-scale supervised learning incomplex visual scenes. In this paper we present practical semi-supervisedand self-supervised models that support training and good generalizationin real-world images and video. Our formulation is based on kinematiclatent normalizing flow representations and dynamics, as well as differ-entiable, semantic body part alignment loss functions that support self-supervised learning. In extensive experiments using 3D motion capturedatasets like CMU, Human3.6M, 3DPW, or AMASS, as well as imagerepositories like COCO, we show that the proposed methods outperformthe state of the art, supporting the practical construction of an accuratefamily of models based on large-scale training with diverse and incom-pletely labeled image and video data. View details
    Preview abstract We present a novel GAN-based model that utilizes the space of deep features learned by a pre-trained object recognition model. Inspired by classical image pyramid representations, we construct our model as a Semantic Generation Pyramid -- a hierarchical framework which leverages the continuum of semantic information encapsulated in such deep features; this ranges from low level information contained in fine features to high level, semantic information contained in deeper features. More specifically, given a set of features extracted from a reference image, our model generates diverse image samples, each with matching features at each semantic level of the recognition model. We demonstrate that our model results in a versatile and flexible framework that can be used in various classic and novel image generation tasks. These include: generating images with a controllable extent of semantic similarity to a reference image, and different manipulation tasks such as semantically-controlled inpainting and compositing; all achieved with the same model, with no further training. View details
    Multi-Plane Program Induction with 3D Box Priors
    Yikai Li
    Jiayuan Mao
    Xiuming Zhang
    Josh Tenenbaum
    Jiajun Wu
    Neural Information Processing Systems (NeurIPS) (2020)
    Preview abstract We consider two important structures in understanding and editing images: modeling regular, program-like texture or patterns in 2D planes, and 3D posing of these planes in the scene. Unlike prior work on image-based program synthesis, which assumes the image contains a single visible 2D plane, we present Box Program Induction (BPI), which infers a program-like scene representation that simultaneously models repeated structure on multiple 2D planes, the 3D position and orientation of the planes, and camera parameters, all from a single image. Our model assumes a box prior, i.e., that the image captures either an inner view or an outer view of a box in 3D. It uses neural networks to infer visual cues such as vanishing points, wireframe lines, or plane segmentations to guide a search-based algorithm to find the program that best explains the image. Such a holistic, structured scene representation enables 3D-aware interactive image editing operations such as inpainting missing pixels, changing camera parameters, and extrapolate the image contents. View details
    Semantic Pyramid for Image Generation
    Assaf Shocher
    Yossi Gandelsman
    Michal Irani
    Proc. IEEE Computer Vision and Pattern Recognition (CVPR) (2020)
    Preview abstract We present a novel GAN-based model that utilizes the space of deep features learned by a pre-trained classification model. Inspired by classical image pyramid representations, we construct our model as a Semantic Generation Pyramid - a hierarchical framework which leverages the continuum of semantic information encapsulated in such deep features; this ranges from low level information contained in fine features to high level, semantic information contained in deeper features. More specifically, given a set of features extracted from a reference image, our model generates diverse image samples, each with matching features at each semantic level of the classification model. We demonstrate that our model results in a versatile and flexible framework that can be used in various classic and novel image generation tasks. These include: generating images with a controllable extent of semantic similarity to a reference image, and different manipulation tasks such as semantically-controlled inpainting and compositing; all achieved with the same model, with no further training. View details
    Preview abstract We wish to automatically predict the "speediness" of moving objects in videos---whether they move faster, at, or slower than their "natural" speed. The core component in our approach is SpeedNet---a novel deep network trained to detect if a video is playing at normal rate, or if it is sped up. SpeedNet is trained on a large corpus of natural videos in a self-supervised manner, without requiring any manual annotations. We show how this single, binary classification network can be used to detect arbitrary rates of speediness of objects. We demonstrate prediction results by SpeedNet on a wide range of videos containing complex natural motions, and examine the visual cues it utilizes for making those predictions. Importantly, we show that through predicting the speed of videos, the model learns a powerful and meaningful space-time representation that goes beyond simple motion cues. We demonstrate how those learned features can boost the performance of self-supervised action recognition, and can be used for video retrieval. Furthermore, we also apply SpeedNet for generating time-varying, adaptive video speedups, which can allow viewers to watch videos faster, but with less of the jittery, unnatural motions typical to videos that are sped up uniformly. View details
    Preview abstract We present a method for retiming people in an ordinary, natural video---manipulating and editing the time in which different motions of individuals in the video occur. We can temporally align different motions, change the speed of certain actions (speeding up/slowing down, or entirely "freezing" people), or "erase" selected people from the video altogether. We achieve these effects computationally via a dedicated learning-based layered video representation, where each frame in the video is decomposed into separate RGBA layers, representing the appearance of different people in the video. A key property of our model is that it not only disentangles the direct motions of each person in the input video, but also correlates each person automatically with the scene changes they generate---e.g., shadows, reflections, and motion of loose clothing. The layers can be individually retimed and recombined into a new video, allowing us to achieve realistic, high-quality renderings of retiming effects for real-world videos depicting complex actions and involving multiple individuals, including dancing, trampoline jumping, or group running. View details
    Learning the Depths of Moving People by Watching Frozen People
    Zhengqi Li
    Ce Liu
    Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2019)
    Preview abstract We present a method for predicting dense depth in scenarios where both a monocular camera and people in the scene are freely moving. Existing methods for recovering depth for dynamic, non-rigid objects from monocular video impose strong assumptions on the objects' motion and often can recover only a sparse depth. In this paper, we take a data-driven approach and learn human depth priors from a large corpus of data. Specifically, we use a new source of data comprised of thousands of Internet videos in which people imitate mannequins, i.e., people freeze in diverse, natural poses, while a hand-held camera is touring the scene. We then create training data using modern Multi-View Stereo (MVS) methods, and design a model that is applied to dynamic scene at inference time. Our method makes use of motion parallax beyond single view and shows clear advantages over state-of-the-art monocular depth prediction methods. We demonstrate the applicability of our method on real-world sequences captured by a moving hand-held camera, depicting complex human actions. We show various 3D effects such as re-focusing, creating a stereoscopic video from a monocular one, and inserting virtual objects to the scene, all produced using our predicted depth maps. View details
    Preview abstract Humans easily recognize object parts and their hierarchical structure by watching how they move; they can then predict how each part moves in the future. In this paper, we propose a novel formulation that simultaneously learns a hierarchical, disentangled object representation and a dynamics model for object parts from unlabeled videos. Our Parts, Structure, and Dynamics (PSD) model learns to, first, recognize the object parts via a layered image representation; second, predict hierarchy via a structural descriptor that composes low-level concepts into a hierarchical structure; and third, model the system dynamics by predicting the future. Experiments on multiple real and synthetic datasets demonstrate that our PSD model works well on all three tasks: segmenting object parts, building their hierarchical structure, and capturing their motion distributions. View details
    Unsupervised Training for 3D Morphable Model Regression
    Kyle Genova
    Aaron Maschinot
    Daniel Vlasic
    The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2018)
    Preview abstract We present a method for training a regression network from image pixels to 3D morphable model coordinates using only unlabeled photographs. The training loss is based on features from a facial recognition network, computed on-the-fly by rendering the predicted faces with a differentiable renderer. To make training from features feasible and avoid network fooling effects, we introduce three objectives: a batch regularization loss that encourages the output distribution to match the distribution of the morphable model, a loopback loss that ensures the regression network can correctly reinterpret its own output, and a multi-view loss that compares the predicted 3D face to the input photograph from multiple viewing angles. We train a regression network using these objectives, a set of unlabeled photographs, and the morphable model itself, and demonstrate state-of-the-art results. View details
    Preview abstract We present a model for isolating and enhancing speech of desired speakers in a video. The input is a video with one or more people speaking, where the speech of interest is interfered by other speakers and/or background noise. We leverage both audio and visual features for this task, which are fed into a joint audio-visual source separation model we designed and trained using thousands of hours of video segments with clean speech from our new dataset, AVSpeech-90K. We present results for various real, practical scenarios involving heated debates and interviews, noisy bars and screaming children, only requiring users to specify the face of the person in the video whose speech they would like to isolate. View details
    MoSculp: Interactive Visualization of Shape and Time
    Andrew Owens
    Jiajun Wu
    Qiurui He
    Tianfan Xue
    Xiuming Zhang
    stefanie mueller
    UIST'18 (2018)
    Preview abstract We present a system that allows users to visualize complex human motion via 3D motion sculptures---a representation that conveys the 3D structure swept by a human body as it moves through space. Given an input video, our system computes a motion sculpture and provides the user with an interface for rendering it in different styles, including the options to insert the sculpture back into the source video or render it in a synthetic scene. To provide this end-to-end workflow, we introduce an algorithm that estimates a human's 3D geometry over time and a 3D-aware image-based rendering approach that preserves the depth ordering of their body motions. By automating the process, our system takes motion sculpture creation out of the realm of professional artists, and makes it applicable to a wide range of existing video material. By providing viewers with 3D information, motion sculptures reveal space-time motion information that is difficult to perceive with the naked eye, and allow viewers to interpret how different parts of the object interact over time. We validate the effectiveness of this approach with user studies, finding that our motion sculpture visualizations are significantly more informative about motion than existing stroboscopic and space-time visualization methods. View details
    Preview abstract We present a joint audio-visual model for isolating a single speech signal from a mixture of sounds such as other speakers and background noise. Solving this task using only audio as input is extremely challenging and does not provide an association of the separated speech signals with speakers in the video. In this paper, we present a deep network-based model that incorporates both visual and auditory signals to solve this task. The visual features are used to "focus" the audio on desired speakers in a scene and to improve the speech separation quality. To train our joint audio-visual model, we introduce AVSpeech, a new dataset comprised of thousands of hours of video segments from the Web. We demonstrate the applicability of our method to classic speech separation tasks, as well as real-world scenarios involving heated interviews, noisy bars, and screaming children, only requiring the user to specify the face of the person in the video whose speech they want to isolate. Our method shows clear advantage over state-of-the-art audio-only speech separation in cases of mixed speech. In addition, our model, which is speaker-independent (trained once, applicable to any speaker), produces better results than recent audio-visual speech separation methods that are speaker-dependent (require training a separate model for each speaker of interest). View details
    Sparse, Smart Contours to Represent and Edit Images
    Ce Liu
    Chuang Gan
    Dilip Krishnan
    Computer Vision and Pattern Recognition (2018)
    Preview abstract We study the problem of reconstructing an image from information stored at sparse contour locations comprising less than $6\%$ of image pixels. This extremely sparse representation provides an intuitive interface for semantically-aware image manipulation. Local edits in contour domain translate to long-range and coherent changes in pixel space. We use generative adversarial networks to synthesize texture and structure even in regions where no input information is provided. With our setup, we can perform complex structural changes such as changing facial expression and interpolating animal fur texture by simple edits of contours such as scaling, moving and erasing. Experiments on a variety of datasets verify the versatility and convenience afforded by our models. View details
    On the Effectiveness of Visible Watermarks
    Ce Liu
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2017)
    Preview abstract Visible watermarking is a widely-used technique for marking and protecting copyrights of many millions of images on the web, yet it suffers from an inherent security flaw—watermarks are typically added in a consistent manner to many images. We show that this consistency allows to automatically estimate the watermark and recover the original images with high accuracy. Specifically, we present a generalized multi-image matting algorithm that takes a watermarked image collection as input and automatically estimates the “foreground” (watermark), its alpha matte, and the “background” (original) images. Since such an attack relies on the consistency of watermarks across image collection, we explore and evaluate how it is affected by various types of inconsistencies in the watermark embedding that could potentially be used to make watermarking more secured. We demonstrate the algorithm on stock imagery available on the web, and provide extensive quantitative analysis on synthetic watermarked data. A key takeaway message of this paper is that visible watermarks should be designed to not only be robust against removal from a single image, but to be more resistant to mass-scale removal from image collections as well. View details
    Preview abstract We present a method for synthesizing a frontal, neutral-expression image of a person's face given an input face photograph. This is achieved by learning to generate facial landmarks and textures from features extracted from a facial-recognition network. Unlike previous approaches, our encoding feature vector is largely invariant to lighting, pose, and facial expression. Exploiting this invariance, we train our decoder network using only frontal, neutral-expression photographs. Since these photographs are well aligned, we can decompose them into a sparse set of landmark points and aligned texture maps. The decoder then predicts landmarks and textures independently and combines them using a differentiable image warping operation. The resulting images can be used for a number of applications, such as analyzing facial attributes, exposure and white balance adjustment, or creating a 3-D avatar. View details
    No Results Found