Figure 1: We study the task of Canonical Surface Mapping (CSM). This task is a generalization of keypoint estimation and involves mapping pixels to canonical 3D models. We learn CSM prediction without requiring correspondence annotations, by instead using geometric cycle consistency as supervision. This allows us to train CSM prediction for diverse classes, including rigid and non-rigid objects. AbstractWe explore the task of Canonical Surface Mapping (CSM). Specifically, given an image, we learn to map pixels on the object to their corresponding locations on an abstract 3D model of the category. But how do we learn such a mapping? A supervised approach would require extensive manual labeling which is not scalable beyond a few hand-picked categories. Our key insight is that the CSM task (pixel to 3D), when combined with 3D projection (3D to pixel), completes a cycle. Hence, we can exploit a geometric cycle consistency loss, thereby allowing us to forgo the dense manual supervision. Our approach allows us to train a CSM model for a diverse set of classes, without sparse or dense keypoint annotation, by leveraging only foreground mask labels for training. We show that our predictions also allow us to infer dense correspondence between two images, and compare the performance of our approach against several methods that predict correspondence by leveraging varying amount of supervision.* the last two authors were equally uninvolved.

No abstract

SUMMARYTreatment of a pathogenic disease process is interpreted as the optimal control of a dynamic system. Evolution of the disease is characterized by a non-linear, fourth-order ordinary differential equation that describes concentrations of pathogens, plasma cells, and antibodies, as well as a numerical indication of patient health. Without control, the dynamic model evidences sub-clinical or clinical decay, chronic stabilization, or unrestrained lethal growth of the pathogen, depending on the initial conditions for the infection. The dynamic equations are controlled by therapeutic agents that affect the rate of change of system variables. Control histories that minimize a quadratic cost function are generated by numerical optimization over a fixed time interval, given otherwise lethal initial conditions. Tradeoffs between cost function weighting of pathogens, organ health, and use of therapeutics are evaluated. Optimal control solutions that defeat the pathogen and preserve organ health are demonstrated for four different approaches to therapy. It is shown that control theory can point the way toward new protocols for treatment and remediation of human diseases.

https://nileshkulkarni.github.io/relative3d/ (b) Results (a) Overview of our method Translation Rotation Scale Instance BBoxes Union (pair) BBox Object Encoder Relative Encoder Final Prediction Image with detections Per object Per pair f Figure 1: (a) Approach Overview: We study the problem of layout estimation in 3D by reasoning about relationships between objects. Given an image and object detection boxes, we first predict the 3D pose (translation, rotation, scale) of each object and the relative pose between each pair of objects. We combine these predictions and ensure consistent relationships between objects to predict the final 3D pose of each object. (b) Output: An example result of our method that takes as input the 2D image and generates the 3D layout. AbstractWe propose an approach to predict the 3D shape and pose for the objects present in a scene. Existing learning based methods that pursue this goal make independent predictions per object, and do not leverage the relationships amongst them. We argue that reasoning about these relationships is crucial, and present an approach to incorporate these in a 3D prediction framework. In addition to independent per-object predictions, we predict pairwise relations in the form of relative 3D pose, and demonstrate that these can be easily incorporated to improve object level estimates. We report performance across different datasets (SUNCG, NYUv2), and show that our approach significantly improves over independent prediction approaches while also outperforming alternate implicit reasoning methods.

stengel@princeton.edu; rghiglia@princeton.edu; nkulkarn@princeton.edu

https://shubhtuls.github.io/imr/ Figure 1: Given a single input image, we can infer the shape, texture and camera viewpoint for the underlying object. In rows 1 and 2, we show the input image, inferred 3D shape and texture from the predicted viewpoint, and three novel viewpoints. We can learn 3D inference using only in-the-wild image collections with approximate instance segmentations, our approach can be easily applied across a diverse set of categories. Rows 3 and 4 show sample predictions across a broad set of categories, with the predicted 3D shape overlaid on the input image. Please see the project page for 360 degree visualizations.

No abstract

Neural networks (NNs) have been successfully used for implementing control architectures for different applications. In this paper, we examine NN augmented intelligent control of a turbo-fan engine toward the goal of minimizing a performance measure on-line. This architecture utilizes an adaptive critic to estimate the engine performance, which is then used to train an NN demand generator for minimizing the performance measure. The present architecture is implemented on a nonlinear model that was provided by General Electric. The model simulates a changed engine by changing the flow and efficiency scalars of the various components of the engine. Results of using the adaptive critic-based performance seeking control architecture show excellent improvement in performance over time.