This paper presents a novel method for entity disambiguation in anonymized graphs using local neighborhood structure. Most existing approaches leverage node information, which might not be available in several contexts due to privacy concerns, or information about the sources of the data. We consider this problem in the supervised setting where we are provided only with a base graph and a set of nodes labelled as ambiguous or unambiguous. We characterize the similarity between two nodes based on their local neighborhood structure using graph kernels; and solve the resulting classification task using SVMs. We give empirical evidence on two real-world datasets, comparing our approach to a state-of-the-art method, highlighting the advantages of our approach. We show that using less information, our method is significantly better in terms of either speed or accuracy or both. We also present extensions of two existing graphs kernels, namely, the direct product kernel and the shortestpath kernel, with significant improvements in accuracy. For the direct product kernel, our extension also provides significant computational benefits. Moreover, we design and implement the algorithms of our method to work in a distributed fashion using the GraphLab framework, ensuring high scalability.

This paper presents a method of estimating the geometry of a room and the 3D pose of objects from a single 360 • panorama image. Assuming Manhattan World geometry, we formulate the task as a Bayesian inference problem in which we estimate positions and orientations of walls and objects. The method combines surface normal estimation, 2D object detection and 3D object pose estimation. Quantitative results are presented on a dataset of synthetically generated 3D rooms containing objects, as well as on a subset of hand-labeled images from the public SUN360 dataset.

Efficient and reliable methods for training of object detectors are in higher demand than ever, and more and more data relevant to the field is becoming available. However, large datasets like Open Images Dataset v4 (OID) are sparsely annotated, and some measure must be taken in order to ensure the training of a reliable detector. In order to take the incompleteness of these datasets into account, one possibility is to use pretrained models to detect the presence of the unverified objects. However, the performance of such a strategy depends largely on the power of the pretrained model. In this study, we propose part-aware sampling, a method that uses human intuition for the hierarchical relation between objects. In terse terms, our method works by making assumptions like "a bounding box for a car should contain a bounding box for a tire". We demonstrate the power of our method on OID and compare the performance against a method based on a pretrained model. Our method also won the first and second place on the public and private test sets of the Google AI Open Images Competition 2018.

We present an approach for road segmentation that only requires image-level annotations at training time. We leverage distant supervision, which allows us to train our model using images that are different from the target domain. Using large publicly available image databases as distant supervisors, we develop a simple method to automatically generate weak pixel-wise road masks. These are used to iteratively train a fully convolutional neural network, which produces our final segmentation model. We evaluate our method on the Cityscapes dataset, where we compare it with a fully supervised approach. Further, we discuss the tradeoff between annotation cost and performance. Overall, our distantly supervised approach achieves 93.8% of the performance of the fully supervised approach, while using orders of magnitude less annotation work.

Identifying "free-space," or safely driveable regions in the scene ahead, is a fundamental task for autonomous navigation. While this task can be addressed using semantic segmentation, the manual labor involved in creating pixelwise annotations to train the segmentation model is very costly. Although weakly supervised segmentation addresses this issue, most methods are not designed for free-space. In this paper, we observe that homogeneous texture and location are two key characteristics of free-space, and develop a novel, practical framework for free-space segmentation with minimal human supervision. Our experiments show that our framework performs better than other weakly supervised methods while using less supervision. Our work demonstrates the potential for performing free-space segmentation without tedious and costly manual annotation, which will be important for adapting autonomous driving systems to different types of vehicles and environments.

No abstract

We present a large-scale object detection system by team PFDet. Our system enables training with huge datasets using 512 GPUs, handles sparsely verified classes, and massive class imbalance. Using our method, we achieved 2nd place in the Google AI Open Images Object Detection Track 2018 on Kaggle. 1 * The authors contributed equally and they are ordered alphabetically.1 https://www.kaggle.com/c/ google-ai-open-images-object-detection-track

We present a method for view-invariant action recognition from depth cameras based on graph signal processing techniques. Our framework leverages a novel graph representation of an action as a temporal sequence of graphs, onto which we apply a spectral graph wavelet transform for creating our feature descriptor. We evaluate two view-invariant graph types: skeleton-based and keypoint-based. The skeleton-based descriptor captures the spatial pose of the subject, whereas the keypoint-based is able to capture complementary information about human-object interaction and the shape of the point cloud. We investigate the effectiveness of our method by experiments on five publicly available datasets. By the graph structure, our method captures the temporal interaction between depth map interest points and achieves a 19.8% increase in performance compared to state-of-the-art results for cross-view action recognition, and competing results for frontal-view action recognition and human-object interaction. Namely, our method results in 90.8% accuracy on the cross-view N-UCLA Multiview Action3D dataset and 91.4% accuracy on the challenging MSRAction3D dataset in the cross-subject setting. For human-object interaction, our method achieves 72.3% accuracy on the Online RGBD Action dataset. We also achieve 96.0% and 98.8% accuracy on the MSRActionPairs3D and UCF-Kinect datasets, respectively.