We introduce Habitat 2.0 (H2.0), a simulation platform for training virtual robots in interactive 3D environments and complex physics-enabled scenarios. We make comprehensive contributions to all levels of the embodied AI stack - data, simulation, and benchmark tasks. Specifically, we present: (i) ReplicaCAD: an artist-authored, annotated, reconfigurable 3D dataset of apartments (matching real spaces) with articulated objects (e.g. cabinets and drawers that can open/close); (ii) H2.0: a high-performance physics-enabled 3D simulator with speeds exceeding 25,000 simulation steps per second (850x real-time) on an 8-GPU node, representing 100x speed-ups over prior work; and, (iii) Home Assistant Benchmark (HAB): a suite of common tasks for assistive robots (tidy the house, prepare groceries, set the table) that test a range of mobile manipulation capabilities. These large-scale engineering contributions allow us to systematically compare deep reinforcement learning (RL) at scale and classical sense-plan-act (SPA) pipelines in long-horizon structured tasks, with an emphasis on generalization to new objects, receptacles, and layouts. We find that (1) flat RL policies struggle on HAB compared to hierarchical ones; (2) a hierarchy with independent skills suffers from 'hand-off problems', and (3) SPA pipelines are more brittle than RL policies.
We present Multiscale Vision Transformers (MViT) for video and image recognition, by connecting the seminal idea of multiscale feature hierarchies with transformer models. Multiscale Transformers have several channel-resolution scale stages. Starting from the input resolution and a small channel dimension, the stages hierarchically expand the channel capacity while reducing the spatial resolution. This creates a multiscale pyramid of features with early layers operating at high spatial resolution to model simple low-level visual information, and deeper layers at spatially coarse, but complex, high-dimensional features. We evaluate this fundamental architectural prior for modeling the dense nature of visual signals for a variety of video recognition tasks where it outperforms concurrent vision transformers that rely on large scale external pre-training and are 5-10x more costly in computation and parameters. We further remove the temporal dimension and apply our model for image classification where it outperforms prior work on vision transformers. Code is available at: https://github.com/facebookresearch/SlowFast
While improving prediction accuracy has been the focus of machine learning in recent years, this alone does not suffice for reliable decision-making. Deploying learning systems in consequential settings also requires calibrating and communicating the uncertainty of predictions. To convey instance-wise uncertainty for prediction tasks, we show how to generate set-valued predictions from a black-box predictor that control the expected loss on future test points at a user-specified level. Our approach provides explicit finite-sample guarantees for any dataset by using a holdout set to calibrate the size of the prediction sets. This framework enables simple, distribution-free, rigorous error control for many tasks, and we demonstrate it in five large-scale machine learning problems: (1) classification problems where some mistakes are more costly than others; (2) multi-label classification, where each observation has multiple associated labels; (3) classification problems where the labels have a hierarchical structure; (4) image segmentation, where we wish to predict a set of pixels containing an object of interest; and (5) protein structure prediction. Lastly, we discuss extensions to uncertainty quantification for ranking, metric learning and distributionally robust learning.
In this work we explore reconstructing hand-object interactions in the wild. The core challenge of this problem is the lack of appropriate 3D labeled data. To overcome this issue, we propose an optimization-based procedure which does not require direct 3D supervision. The general strategy we adopt is to exploit all available related data (2D bounding boxes, 2D hand keypoints, 2D instance masks, 3D object models, 3D in-the-lab MoCap) to provide constraints for the 3D reconstruction. Rather than optimizing the hand and object individually, we optimize them jointly which allows us to impose additional constraints based on hand-object contact, collision, and occlusion. Our method produces compelling reconstructions on the challenging in-the-wild data from the EPIC Kitchens and the 100 Days of Hands datasets, across a range of object categories. Quantitatively, we demonstrate that our approach compares favorably to existing approaches in the lab settings where ground truth 3D annotations are available.
Videos from edited media like movies are a useful, yet under-explored source of information. The rich variety of appearance and interactions between humans depicted over a large temporal context in these films could be a valuable source of data. However, the richness of data comes at the expense of fundamental challenges such as abrupt shot changes and close up shots of actors with heavy truncation, which limits the applicability of existing human 3D understanding methods. In this paper, we address these limitations with an insight that while shot changes of the same scene incur a discontinuity between frames, the 3D structure of the scene still changes smoothly. This allows us to handle frames before and after the shot change as multi-view signal that provide strong cues to recover the 3D state of the actors. We propose a multi-shot optimization framework, which leads to improved 3D reconstruction and mining of long sequences with pseudo ground truth 3D human mesh. We show that the resulting data is beneficial in the training of various human mesh recovery models: for single image, we achieve improved robustness; for video we propose a pure transformer-based temporal encoder, which can naturally handle missing observations due to shot changes in the input frames. We demonstrate the importance of the insight and proposed models through extensive experiments. The tools we develop open the door to processing and analyzing in 3D content from a large library of edited media, which could be helpful for many downstream applications. Project page: https://geopavlakos.github.io/multishot
Human trajectory forecasting is an inherently multi-modal problem. Uncertainty in future trajectories stems from two sources: (a) sources that are known to the agent but unknown to the model, such as long term goals and (b)sources that are unknown to both the agent & the model, such as intent of other agents & irreducible randomness indecisions. We propose to factorize this uncertainty into its epistemic & aleatoric sources. We model the epistemic un-certainty through multimodality in long term goals and the aleatoric uncertainty through multimodality in waypoints& paths. To exemplify this dichotomy, we also propose a novel long term trajectory forecasting setting, with prediction horizons upto a minute, an order of magnitude longer than prior works. Finally, we presentY-net, a scene com-pliant trajectory forecasting network that exploits the pro-posed epistemic & aleatoric structure for diverse trajectory predictions across long prediction horizons.Y-net significantly improves previous state-of-the-art performance on both (a) The well studied short prediction horizon settings on the Stanford Drone & ETH/UCY datasets and (b) The proposed long prediction horizon setting on the re-purposed Stanford Drone & Intersection Drone datasets.
In continual learning, new categories may be introduced over time, and an ideal learning system should perform well on both the original categories and the new categories. While deep neural nets have achieved resounding success in the classical supervised setting, they are known to forget about knowledge acquired in prior episodes of learning if the examples encountered in the current episode of learning are drastically different from those encountered in prior episodes. In this paper, we propose a new method that can both leverage the expressive power of deep neural nets and is resilient to forgetting when new categories are introduced. We found the proposed method can reduce forgetting by 2.3x to 6.9x on CIFAR-10 compared to existing methods and by 1.8x to 2.7x on ImageNet compared to an oracle baseline.
Vision-based robotics often separates the control loop into one module for perception and a separate module for control. It is possible to train the whole system end-to-end (e.g. with deep RL), but doing it "from scratch" comes with a high sample complexity cost and the final result is often brittle, failing unexpectedly if the test environment differs from that of training. We study the effects of using mid-level visual representations (features learned asynchronously for traditional computer vision objectives), as a generic and easy-to-decode perceptual state in an end-to-end RL framework. Mid-level representations encode invariances about the world, and we show that they aid generalization, improve sample complexity, and lead to a higher final performance. Compared to other approaches for incorporating invariances, such as domain randomization, asynchronously trained mid-level representations scale better: both to harder problems and to larger domain shifts. In practice, this means that mid-level representations could be used to successfully train policies for tasks where domain randomization and learning-from-scratch failed. We report results on both manipulation and navigation tasks, and for navigation include zero-shot sim-to-real experiments on real robots.
We describe a framework for research and evaluation in Embodied AI. Our proposal is based on a canonical task: Rearrangement. A standard task can focus the development of new techniques and serve as a source of trained models that can be transferred to other settings. In the rearrangement task, the goal is to bring a given physical environment into a specified state. The goal state can be specified by object poses, by images, by a description in language, or by letting the agent experience the environment in the goal state. We characterize rearrangement scenarios along different axes and describe metrics for benchmarking rearrangement performance. To facilitate research and exploration, we present experimental testbeds of rearrangement scenarios in four different simulation environments. We anticipate that other datasets will be released and new simulation platforms will be built to support training of rearrangement agents and their deployment on physical systems.
A fundamental problem in computer vision is that of inferring the intrinsic, 3D structure of the world from flat, 2D images of that world. Traditional methods for recovering scene properties such as shape, reflectance, or illumination rely on multiple observations of the same scene to overconstrain the problem. Recovering these same properties from a single image seems almost impossible in comparison -- there are an infinite number of shapes, paint, and lights that exactly reproduce a single image. However, certain explanations are more likely than others: surfaces tend to be smooth, paint tends to be uniform, and illumination tends to be natural. We therefore pose this problem as one of statistical inference, and define an optimization problem that searches for the *most likely* explanation of a single image. Our technique can be viewed as a superset of several classic computer vision problems (shape-from-shading, intrinsic images, color constancy, illumination estimation, etc) and outperforms all previous solutions to those constituent problems.