Header logo is ps


2020


STAR: Sparse Trained Articulated Human Body Regressor
STAR: Sparse Trained Articulated Human Body Regressor

Osman, A. A. A., Bolkart, T., Black, M. J.

In European Conference on Computer Vision (ECCV) , August 2020 (inproceedings)

Abstract
The SMPL body model is widely used for the estimation, synthesis, and analysis of 3D human pose and shape. While popular, we show that SMPL has several limitations and introduce STAR, which is quantitatively and qualitatively superior to SMPL. First, SMPL has a huge number of parameters resulting from its use of global blend shapes. These dense pose-corrective offsets relate every vertex on the mesh to all the joints in the kinematic tree, capturing spurious long-range correlations. To address this, we define per-joint pose correctives and learn the subset of mesh vertices that are influenced by each joint movement. This sparse formulation results in more realistic deformations and significantly reduces the number of model parameters to 20% of SMPL. When trained on the same data as SMPL, STAR generalizes better despite having many fewer parameters. Second, SMPL factors pose-dependent deformations from body shape while, in reality, people with different shapes deform differently. Consequently, we learn shape-dependent pose-corrective blend shapes that depend on both body pose and BMI. Third, we show that the shape space of SMPL is not rich enough to capture the variation in the human population. We address this by training STAR with an additional 10,000 scans of male and female subjects, and show that this results in better model generalization. STAR is compact, generalizes better to new bodies and is a drop-in replacement for SMPL. STAR is publicly available for research purposes at http://star.is.tue.mpg.de.

Project Page Code Video paper supplemental [BibTex]


Monocular Expressive Body Regression through Body-Driven Attention
Monocular Expressive Body Regression through Body-Driven Attention

Choutas, V., Pavlakos, G., Bolkart, T., Tzionas, D., Black, M. J.

In Computer Vision – ECCV 2020, Springer International Publishing, Cham, August 2020 (inproceedings)

Abstract
To understand how people look, interact, or perform tasks,we need to quickly and accurately capture their 3D body, face, and hands together from an RGB image. Most existing methods focus only on parts of the body. A few recent approaches reconstruct full expressive 3D humans from images using 3D body models that include the face and hands. These methods are optimization-based and thus slow, prone to local optima, and require 2D keypoints as input. We address these limitations by introducing ExPose (EXpressive POse and Shape rEgression), which directly regresses the body, face, and hands, in SMPL-X format, from an RGB image. This is a hard problem due to the high dimensionality of the body and the lack of expressive training data. Additionally, hands and faces are much smaller than the body, occupying very few image pixels. This makes hand and face estimation hard when body images are downscaled for neural networks. We make three main contributions. First, we account for the lack of training data by curating a dataset of SMPL-X fits on in-the-wild images. Second, we observe that body estimation localizes the face and hands reasonably well. We introduce body-driven attention for face and hand regions in the original image to extract higher-resolution crops that are fed to dedicated refinement modules. Third, these modules exploit part-specific knowledge from existing face and hand-only datasets. ExPose estimates expressive 3D humans more accurately than existing optimization methods at a small fraction of the computational cost. Our data, model and code are available for research at https://expose.is.tue.mpg.de.

code Short video Long video arxiv pdf suppl link (url) Project Page [BibTex]


GRAB: A Dataset of Whole-Body Human Grasping of Objects
GRAB: A Dataset of Whole-Body Human Grasping of Objects

Taheri, O., Ghorbani, N., Black, M. J., Tzionas, D.

In Computer Vision – ECCV 2020, Springer International Publishing, Cham, August 2020 (inproceedings)

Abstract
Training computers to understand, model, and synthesize human grasping requires a rich dataset containing complex 3D object shapes, detailed contact information, hand pose and shape, and the 3D body motion over time. While "grasping" is commonly thought of as a single hand stably lifting an object, we capture the motion of the entire body and adopt the generalized notion of "whole-body grasps". Thus, we collect a new dataset, called GRAB (GRasping Actions with Bodies), of whole-body grasps, containing full 3D shape and pose sequences of 10 subjects interacting with 51 everyday objects of varying shape and size. Given MoCap markers, we fit the full 3D body shape and pose, including the articulated face and hands, as well as the 3D object pose. This gives detailed 3D meshes over time, from which we compute contact between the body and object. This is a unique dataset, that goes well beyond existing ones for modeling and understanding how humans grasp and manipulate objects, how their full body is involved, and how interaction varies with the task. We illustrate the practical value of GRAB with an example application; we train GrabNet, a conditional generative network, to predict 3D hand grasps for unseen 3D object shapes. The dataset and code are available for research purposes at https://grab.is.tue.mpg.de.

pdf suppl video (long) video (short) link (url) DOI [BibTex]

pdf suppl video (long) video (short) link (url) DOI [BibTex]


Learning to Dress 3D People in Generative Clothing
Learning to Dress 3D People in Generative Clothing

Ma, Q., Yang, J., Ranjan, A., Pujades, S., Pons-Moll, G., Tang, S., Black, M. J.

In Computer Vision and Pattern Recognition (CVPR), pages: 6468-6477, IEEE, June 2020 (inproceedings)

Abstract
Three-dimensional human body models are widely used in the analysis of human pose and motion. Existing models, however, are learned from minimally-clothed 3D scans and thus do not generalize to the complexity of dressed people in common images and videos. Additionally, current models lack the expressive power needed to represent the complex non-linear geometry of pose-dependent clothing shape. To address this, we learn a generative 3D mesh model of clothed people from 3D scans with varying pose and clothing. Specifically, we train a conditional Mesh-VAE-GAN to learn the clothing deformation from the SMPL body model, making clothing an additional term on SMPL. Our model is conditioned on both pose and clothing type, giving the ability to draw samples of clothing to dress different body shapes in a variety of styles and poses. To preserve wrinkle detail, our Mesh-VAE-GAN extends patchwise discriminators to 3D meshes. Our model, named CAPE, represents global shape and fine local structure, effectively extending the SMPL body model to clothing. To our knowledge, this is the first generative model that directly dresses 3D human body meshes and generalizes to different poses.

Project page Code Short video Long video arXiv DOI [BibTex]

Project page Code Short video Long video arXiv DOI [BibTex]


{GENTEL : GENerating Training data Efficiently for Learning to segment medical images}
GENTEL : GENerating Training data Efficiently for Learning to segment medical images

Thakur, R. P., Rocamora, S. P., Goel, L., Pohmann, R., Machann, J., Black, M. J.

Congrès Reconnaissance des Formes, Image, Apprentissage et Perception (RFAIP), June 2020 (conference)

Abstract
Accurately segmenting MRI images is crucial for many clinical applications. However, manually segmenting images with accurate pixel precision is a tedious and time consuming task. In this paper we present a simple, yet effective method to improve the efficiency of the image segmentation process. We propose to transform the image annotation task into a binary choice task. We start by using classical image processing algorithms with different parameter values to generate multiple, different segmentation masks for each input MRI image. Then, instead of segmenting the pixels of the images, the user only needs to decide whether a segmentation is acceptable or not. This method allows us to efficiently obtain high quality segmentations with minor human intervention. With the selected segmentations, we train a state-of-the-art neural network model. For the evaluation, we use a second MRI dataset (1.5T Dataset), acquired with a different protocol and containing annotations. We show that the trained network i) is able to automatically segment cases where none of the classical methods obtain a high quality result ; ii) generalizes to the second MRI dataset, which was acquired with a different protocol and was never seen at training time ; and iii) enables detection of miss-annotations in this second dataset. Quantitatively, the trained network obtains very good results: DICE score - mean 0.98, median 0.99- and Hausdorff distance (in pixels) - mean 4.7, median 2.0-.

Project Page PDF [BibTex]

Project Page PDF [BibTex]


Generating 3D People in Scenes without People
Generating 3D People in Scenes without People

Zhang, Y., Hassan, M., Neumann, H., Black, M. J., Tang, S.

In Computer Vision and Pattern Recognition (CVPR), pages: 6194-6204, June 2020 (inproceedings)

Abstract
We present a fully automatic system that takes a 3D scene and generates plausible 3D human bodies that are posed naturally in that 3D scene. Given a 3D scene without people, humans can easily imagine how people could interact with the scene and the objects in it. However, this is a challenging task for a computer as solving it requires that (1) the generated human bodies to be semantically plausible within the 3D environment (e.g. people sitting on the sofa or cooking near the stove), and (2) the generated human-scene interaction to be physically feasible such that the human body and scene do not interpenetrate while, at the same time, body-scene contact supports physical interactions. To that end, we make use of the surface-based 3D human model SMPL-X. We first train a conditional variational autoencoder to predict semantically plausible 3D human poses conditioned on latent scene representations, then we further refine the generated 3D bodies using scene constraints to enforce feasible physical interaction. We show that our approach is able to synthesize realistic and expressive 3D human bodies that naturally interact with 3D environment. We perform extensive experiments demonstrating that our generative framework compares favorably with existing methods, both qualitatively and quantitatively. We believe that our scene-conditioned 3D human generation pipeline will be useful for numerous applications; e.g. to generate training data for human pose estimation, in video games and in VR/AR. Our project page for data and code can be seen at: \url{https://vlg.inf.ethz.ch/projects/PSI/}.

Code PDF DOI [BibTex]

Code PDF DOI [BibTex]


Learning Physics-guided Face Relighting under Directional Light
Learning Physics-guided Face Relighting under Directional Light

Nestmeyer, T., Lalonde, J., Matthews, I., Lehrmann, A. M.

In Conference on Computer Vision and Pattern Recognition, pages: 5123-5132, IEEE/CVF, June 2020 (inproceedings) Accepted

Abstract
Relighting is an essential step in realistically transferring objects from a captured image into another environment. For example, authentic telepresence in Augmented Reality requires faces to be displayed and relit consistent with the observer's scene lighting. We investigate end-to-end deep learning architectures that both de-light and relight an image of a human face. Our model decomposes the input image into intrinsic components according to a diffuse physics-based image formation model. We enable non-diffuse effects including cast shadows and specular highlights by predicting a residual correction to the diffuse render. To train and evaluate our model, we collected a portrait database of 21 subjects with various expressions and poses. Each sample is captured in a controlled light stage setup with 32 individual light sources. Our method creates precise and believable relighting results and generalizes to complex illumination conditions and challenging poses, including when the subject is not looking straight at the camera.

Paper [BibTex]

Paper [BibTex]


{VIBE}: Video Inference for Human Body Pose and Shape Estimation
VIBE: Video Inference for Human Body Pose and Shape Estimation

Kocabas, M., Athanasiou, N., Black, M. J.

In Proceedings IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), pages: 5252-5262, IEEE, June 2020 (inproceedings)

Abstract
Human motion is fundamental to understanding behavior. Despite progress on single-image 3D pose and shape estimation, existing video-based state-of-the-art methodsfail to produce accurate and natural motion sequences due to a lack of ground-truth 3D motion data for training. To address this problem, we propose “Video Inference for Body Pose and Shape Estimation” (VIBE), which makes use of an existing large-scale motion capture dataset (AMASS) together with unpaired, in-the-wild, 2D keypoint annotations. Our key novelty is an adversarial learning framework that leverages AMASS to discriminate between real human motions and those produced by our temporal pose and shape regression networks. We define a temporal network architecture and show that adversarial training, at the sequence level, produces kinematically plausible motion sequences without in-the-wild ground-truth 3D labels. We perform extensive experimentation to analyze the importance of motion and demonstrate the effectiveness of VIBE on challenging 3D pose estimation datasets, achieving state-of-the-art performance. Code and pretrained models are available at https://github.com/mkocabas/VIBE

arXiv code video supplemental video DOI Project Page [BibTex]

arXiv code video supplemental video DOI Project Page [BibTex]


From Variational to Deterministic Autoencoders
From Variational to Deterministic Autoencoders

Ghosh*, P., Sajjadi*, M. S. M., Vergari, A., Black, M. J., Schölkopf, B.

8th International Conference on Learning Representations (ICLR) , April 2020, *equal contribution (conference)

Abstract
Variational Autoencoders (VAEs) provide a theoretically-backed framework for deep generative models. However, they often produce “blurry” images, which is linked to their training objective. Sampling in the most popular implementation, the Gaussian VAE, can be interpreted as simply injecting noise to the input of a deterministic decoder. In practice, this simply enforces a smooth latent space structure. We challenge the adoption of the full VAE framework on this specific point in favor of a simpler, deterministic one. Specifically, we investigate how substituting stochasticity with other explicit and implicit regularization schemes can lead to a meaningful latent space without having to force it to conform to an arbitrarily chosen prior. To retrieve a generative mechanism for sampling new data points, we propose to employ an efficient ex-post density estimation step that can be readily adopted both for the proposed deterministic autoencoders as well as to improve sample quality of existing VAEs. We show in a rigorous empirical study that regularized deterministic autoencoding achieves state-of-the-art sample quality on the common MNIST, CIFAR-10 and CelebA datasets.

arXiv link (url) [BibTex]

arXiv link (url) [BibTex]


Chained Representation Cycling: Learning to Estimate 3D Human Pose and Shape by Cycling Between Representations
Chained Representation Cycling: Learning to Estimate 3D Human Pose and Shape by Cycling Between Representations

Rueegg, N., Lassner, C., Black, M. J., Schindler, K.

In Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-20), pages: 5561-5569, Febuary 2020 (inproceedings)

Abstract
The goal of many computer vision systems is to transform image pixels into 3D representations. Recent popular models use neural networks to regress directly from pixels to 3D object parameters. Such an approach works well when supervision is available, but in problems like human pose and shape estimation, it is difficult to obtain natural images with 3D ground truth. To go one step further, we propose a new architecture that facilitates unsupervised, or lightly supervised, learning. The idea is to break the problem into a series of transformations between increasingly abstract representations. Each step involves a cycle designed to be learnable without annotated training data, and the chain of cycles delivers the final solution. Specifically, we use 2D body part segments as an intermediate representation that contains enough information to be lifted to 3D, and at the same time is simple enough to be learned in an unsupervised way. We demonstrate the method by learning 3D human pose and shape from un-paired and un-annotated images. We also explore varying amounts of paired data and show that cycling greatly alleviates the need for paired data. While we present results for modeling humans, our formulation is general and can be applied to other vision problems.

pdf [BibTex]

pdf [BibTex]

2003


Image statistics and anisotropic diffusion
Image statistics and anisotropic diffusion

Scharr, H., Black, M. J., Haussecker, H.

In Int. Conf. on Computer Vision, pages: 840-847, October 2003 (inproceedings)

pdf [BibTex]

2003

pdf [BibTex]


A switching {Kalman} filter model for the motor cortical coding of hand motion
A switching Kalman filter model for the motor cortical coding of hand motion

Wu, W., Black, M. J., Mumford, D., Gao, Y., Bienenstock, E., Donoghue, J. P.

In Proc. IEEE Engineering in Medicine and Biology Society, pages: 2083-2086, September 2003 (inproceedings)

pdf [BibTex]

pdf [BibTex]


no image
A Gaussian mixture model for the motor cortical coding of hand motion

Wu, W., Mumford, D., Black, M. J., Gao, Y., Bienenstock, E., Donoghue, J. P.

Neural Control of Movement, Santa Barbara, CA, April 2003 (conference)

abstract [BibTex]

abstract [BibTex]


Connecting brains with machines: The neural control of {2D} cursor movement
Connecting brains with machines: The neural control of 2D cursor movement

Black, M. J., Bienenstock, E., Donoghue, J. P., Serruya, M., Wu, W., Gao, Y.

In 1st International IEEE/EMBS Conference on Neural Engineering, pages: 580-583, Capri, Italy, March 2003 (inproceedings)

pdf [BibTex]

pdf [BibTex]


A quantitative comparison of linear and non-linear models of motor cortical activity for the encoding and decoding of arm motions
A quantitative comparison of linear and non-linear models of motor cortical activity for the encoding and decoding of arm motions

Gao, Y., Black, M. J., Bienenstock, E., Wu, W., Donoghue, J. P.

In 1st International IEEE/EMBS Conference on Neural Engineering, pages: 189-192, Capri, Italy, March 2003 (inproceedings)

pdf [BibTex]

pdf [BibTex]


no image
Accuracy of manual spike sorting: Results for the Utah intracortical array

Wood, F., Fellows, M., Vargas-Irwin, C., Black, M. J., Donoghue, J. P.

Program No. 279.2. 2003, Abstract Viewer and Itinerary Planner, Society for Neuroscience, Washington, DC, 2003, Online (conference)

abstract [BibTex]

abstract [BibTex]


no image
Specular flow and the perception of surface reflectance

Roth, S., Domini, F., Black, M. J.

Journal of Vision, 3 (9): 413a, 2003 (conference)

abstract poster [BibTex]

abstract poster [BibTex]


Attractive people: Assembling loose-limbed models using non-parametric belief propagation
Attractive people: Assembling loose-limbed models using non-parametric belief propagation

Sigal, L., Isard, M. I., Sigelman, B. H., Black, M. J.

In Advances in Neural Information Processing Systems 16, NIPS, pages: 1539-1546, (Editors: S. Thrun and L. K. Saul and B. Schölkopf), MIT Press, 2003 (inproceedings)

Abstract
The detection and pose estimation of people in images and video is made challenging by the variability of human appearance, the complexity of natural scenes, and the high dimensionality of articulated body models. To cope with these problems we represent the 3D human body as a graphical model in which the relationships between the body parts are represented by conditional probability distributions. We formulate the pose estimation problem as one of probabilistic inference over a graphical model where the random variables correspond to the individual limb parameters (position and orientation). Because the limbs are described by 6-dimensional vectors encoding pose in 3-space, discretization is impractical and the random variables in our model must be continuous-valued. To approximate belief propagation in such a graph we exploit a recently introduced generalization of the particle filter. This framework facilitates the automatic initialization of the body-model from low level cues and is robust to occlusion of body parts and scene clutter.

pdf (color) pdf (black and white) [BibTex]

pdf (color) pdf (black and white) [BibTex]


Neural decoding of cursor motion using a {Kalman} filter
Neural decoding of cursor motion using a Kalman filter

(Nominated: Best student paper)

Wu, W., Black, M. J., Gao, Y., Bienenstock, E., Serruya, M., Shaikhouni, A., Donoghue, J. P.

In Advances in Neural Information Processing Systems 15, pages: 133-140, MIT Press, 2003 (inproceedings)

pdf [BibTex]

pdf [BibTex]

1999


Edges as outliers: Anisotropic smoothing using local image statistics
Edges as outliers: Anisotropic smoothing using local image statistics

Black, M. J., Sapiro, G.

In Scale-Space Theories in Computer Vision, Second Int. Conf., Scale-Space ’99, pages: 259-270, LNCS 1682, Springer, Corfu, Greece, September 1999 (inproceedings)

Abstract
Edges are viewed as statistical outliers with respect to local image gradient magnitudes. Within local image regions we compute a robust statistical measure of the gradient variation and use this in an anisotropic diffusion framework to determine a spatially varying "edge-stopping" parameter σ. We show how to determine this parameter for two edge-stopping functions described in the literature (Perona-Malik and the Tukey biweight). Smoothing of the image is related the local texture and in regions of low texture, small gradient values may be treated as edges whereas in regions of high texture, large gradient magnitudes are necessary before an edge is preserved. Intuitively these results have similarities with human perceptual phenomena such as masking and "popout". Results are shown on a variety of standard images.

pdf [BibTex]

1999

pdf [BibTex]


Probabilistic detection and tracking of motion discontinuities
Probabilistic detection and tracking of motion discontinuities

(Marr Prize, Honorable Mention)

Black, M. J., Fleet, D. J.

In Int. Conf. on Computer Vision, ICCV-99, pages: 551-558, ICCV, Corfu, Greece, September 1999 (inproceedings)

pdf [BibTex]

pdf [BibTex]


Explaining optical flow events with parameterized spatio-temporal models
Explaining optical flow events with parameterized spatio-temporal models

Black, M. J.

In IEEE Proc. Computer Vision and Pattern Recognition, CVPR’99, pages: 326-332, IEEE, Fort Collins, CO, 1999 (inproceedings)

pdf video [BibTex]

pdf video [BibTex]

1994


Estimating multiple independent motions in segmented images using parametric models with local deformations
Estimating multiple independent motions in segmented images using parametric models with local deformations

Black, M. J., Jepson, A.

In Workshop on Non-rigid and Articulate Motion, pages: 220-227, Austin, Texas, November 1994 (inproceedings)

pdf abstract [BibTex]

1994

pdf abstract [BibTex]


Time to contact from active tracking of motion boundaries
Time to contact from active tracking of motion boundaries

Ju, X., Black, M. J.

In Intelligent Robots and Computer Vision XIII: 3D Vision, Product Inspection, and Active Vision, pages: 26-37, Proc. SPIE 2354, Boston, Massachusetts, November 1994 (inproceedings)

pdf abstract [BibTex]

pdf abstract [BibTex]


The outlier process: Unifying line processes and robust statistics
The outlier process: Unifying line processes and robust statistics

Black, M., Rangarajan, A.

In IEEE Conf. on Computer Vision and Pattern Recognition, CVPR’94, pages: 15-22, Seattle, WA, June 1994 (inproceedings)

pdf abstract [BibTex]

pdf abstract [BibTex]


Recursive non-linear estimation of discontinuous flow fields
Recursive non-linear estimation of discontinuous flow fields

Black, M.

In Proc. Third European Conf. on Computer Vision, ECCV’94,, pages: 138-145, LNCS 800, Springer Verlag, Sweden, May 1994 (inproceedings)

pdf abstract [BibTex]

pdf abstract [BibTex]

1990


A model for the detection of motion over time
A model for the detection of motion over time

Black, M. J., Anandan, P.

In Proc. Int. Conf. on Computer Vision, ICCV-90, pages: 33-37, Osaka, Japan, December 1990 (inproceedings)

Abstract
We propose a model for the recovery of visual motion fields from image sequences. Our model exploits three constraints on the motion of a patch in the environment: i) Data Conservation: the intensity structure corresponding to an environmental surface patch changes gradually over time; ii) Spatial Coherence: since surfaces have spatial extent neighboring points have similar motions; iii) Temporal Coherence: the direction and velocity of motion for a surface patch changes gradually. The formulation of the constraints takes into account the possibility of multiple motions at a particular location. We also present a highly parallel computational model for realizing these constraints in which computation occurs locally, knowledge about the motion increases over time, and occlusion and disocclusion boundaries are estimated. An implementation of the model using a stochastic temporal updating scheme is described. Experiments with both synthetic and real imagery are presented.

pdf [BibTex]

1990

pdf [BibTex]


Constraints for the early detection of discontinuity from motion
Constraints for the early detection of discontinuity from motion

Black, M. J., Anandan, P.

In Proc. National Conf. on Artificial Intelligence, AAAI-90, pages: 1060-1066, Boston, MA, 1990 (inproceedings)

Abstract
Surface discontinuities are detected in a sequence of images by exploiting physical constraints at early stages in the processing of visual motion. To achieve accurate early discontinuity detection we exploit five physical constraints on the presence of discontinuities: i) the shape of the sum of squared differences (SSD) error surface in the presence of surface discontinuities; ii) the change in the shape of the SSD surface due to relative surface motion; iii) distribution of optic flow in a neighborhood of a discontinuity; iv) spatial consistency of discontinuities; V) temporal consistency of discontinuities. The constraints are described, and experimental results on sequences of real and synthetic images are presented. The work has applications in the recovery of environmental structure from motion and in the generation of dense optic flow fields.

pdf [BibTex]

pdf [BibTex]