Watch-n-Patch: Unsupervised Learning of Actions and Relations.

There is a large variation in the activities that humans perform in their everyday lives. We consider modeling these composite human activities which comprises multiple basic level actions in a completely unsupervised setting. Our model learns high-level co-occurrence and temporal relations between the actions. We consider the video as a sequence of short-term action clips, which contains human-words and object-words. An activity is about a set of action-topics and object-topics indicating which actions are present and which objects are interacting with. We then propose a new probabilistic model relating the words and the topics. It allows us to model long-range action relations that commonly exist in the composite activities, which is challenging in previous works. We apply our model to the unsupervised action segmentation and clustering, and to a novel application that detects forgotten actions, which we call action patching. For evaluation, we contribute a new challenging RGB-D activity video dataset recorded by the new Kinect v2, which contains several human daily activities as compositions of multiple actions interacting with different objects. Moreover, we develop a robotic system that watches and reminds people using our action patching algorithm. Our robotic setup can be easily deployed on any assistive robots.

Modeling 3D Environments through Hidden Human Context.

The idea of modeling object-object relations has been widely leveraged in many scene understanding applications. However, as the objects are designed by humans and for human usage, when we reason about a human environment, we reason about it through an interplay between the environment, objects and humans. In this paper, we model environments not only through objects, but also through latent human poses and human-object interactions. In order to handle the large number of latent human poses and a large variety of their interactions with objects, we present Infinite Latent Conditional Random Field (ILCRF) that models a scene as a mixture of CRFs generated from Dirichlet processes. In each CRF, we model objects and object-object relations as existing nodes and edges, and hidden human poses and human-object relations as latent nodes and edges. ILCRF generatively models the distribution of different CRF structures over these latent nodes and edges. We apply the model to the challenging applications of 3D scene labeling and robotic scene arrangement. In extensive experiments, we show that our model significantly outperforms the state-of-the-art results in both applications. We further use our algorithm on a robot for arranging objects in a new scene using the two applications aforementioned.

Anticipating Human Activities Using Object Affordances for Reactive Robotic Response.

An important aspect of human perception is anticipation, which we use extensively in our day-to-day activities when interacting with other humans as well as with our surroundings. Anticipating which activities will a human do next (and how) can enable an assistive robot to plan ahead for reactive responses. Furthermore, anticipation can even improve the detection accuracy of past activities. The challenge, however, is two-fold: We need to capture the rich context for modeling the activities and object affordances, and we need to anticipate the distribution over a large space of future human activities. In this work, we represent each possible future using an anticipatory temporal conditional random field (ATCRF) that models the rich spatial-temporal relations through object affordances. We then consider each ATCRF as a particle and represent the distribution over the potential futures using a set of particles. In extensive evaluation on CAD-120 human activity RGB-D dataset, we first show that anticipation improves the state-of-the-art detection results. We then show that for new subjects (not seen in the training set), we obtain an activity anticipation accuracy (defined as whether one of top three predictions actually happened) of 84.1, 74.4 and 62.2 percent for an anticipation time of 1, 3 and 10 seconds respectively. Finally, we also show a robot using our algorithm for performing a few reactive responses.

3D Reasoning from Blocks to Stability.

Objects occupy physical space and obey physical laws. To truly understand a scene, we must reason about the space that objects in it occupy, and how each objects is supported stably by each other. In other words, we seek to understand which objects would, if moved, cause other objects to fall. This 3D volumetric reasoning is important for many scene understanding tasks, ranging from segmentation of objects to perception of a rich 3D, physically well-founded, interpretations of the scene. In this paper, we propose a new algorithm to parse a single RGB-D image with 3D block units while jointly reasoning about the segments, volumes, supporting relationships, and object stability. Our algorithm is based on the intuition that a good 3D representation of the scene is one that fits the depth data well, and is a stable, self-supporting arrangement of objects (i.e., one that does not topple). We design an energy function for representing the quality of the block representation based on these properties. Our algorithm fits 3D blocks to the depth values corresponding to image segments, and iteratively optimizes the energy function. Our proposed algorithm is the first to consider stability of objects in complex arrangements for reasoning about the underlying structure of the scene. Experimental results show that our stability-reasoning framework improves RGB-D segmentation and scene volumetric representation.

Toward Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models.

Scene understanding includes many related subtasks, such as scene categorization, depth estimation, object detection, etc. Each of these subtasks is often notoriously hard, and state-of-the-art classifiers already exist for many of them. These classifiers operate on the same raw image and provide correlated outputs. It is desirable to have an algorithm that can capture such correlation without requiring any changes to the inner workings of any classifier. We propose Feedback Enabled Cascaded Classification Models (FE-CCM), that jointly optimizes all the subtasks while requiring only a "black box" interface to the original classifier for each subtask. We use a two-layer cascade of classifiers, which are repeated instantiations of the original ones, with the output of the first layer fed into the second layer as input. Our training method involves a feedback step that allows later classifiers to provide earlier classifiers information about which error modes to focus on. We show that our method significantly improves performance in all the subtasks in the domain of scene understanding, where we consider depth estimation, scene categorization, event categorization, object detection, geometric labeling, and saliency detection. Our method also improves performance in two robotic applications: an object-grasping robot and an object-finding robot.

Make3D: learning 3D scene structure from a single still image.

We consider the problem of estimating detailed 3D structure from a single still image of an unstructured environment. Our goal is to create 3D models that are both quantitatively accurate as well as visually pleasing. For each small homogeneous patch in the image, we use a Markov Random Field (MRF) to infer a set of "plane parameters" that capture both the 3D location and 3D orientation of the patch. The MRF, trained via supervised learning, models both image depth cues as well as the relationships between different parts of the image. Other than assuming that the environment is made up of a number of small planes, our model makes no explicit assumptions about the structure of the scene; this enables the algorithm to capture much more detailed 3D structure than does prior art and also give a much richer experience in the 3D flythroughs created using image-based rendering, even for scenes with significant nonvertical structure. Using this approach, we have created qualitatively correct 3D models for 64.9 percent of 588 images downloaded from the Internet. We have also extended our model to produce large-scale 3D models from a few images.