Attributed networks are a type of graph structured data used in many real-world scenarios. Detecting anomalies on attributed networks has a wide spectrum of applications such as spammer detection and fraud detection. Although this research area draws increasing attention in the last few years, previous works are mostly unsupervised because of expensive costs of labeling ground truth anomalies. Many recent studies have shown different types of anomalies are often mixed together on attributed networks and such invaluable human knowledge could provide complementary insights in advancing anomaly detection on attributed networks. To this end, we study the novel problem of modeling and integrating human knowledge of different anomaly types for attributed network anomaly detection. Specifically, we first model prior human knowledge through a novel data augmentation strategy. We then integrate the modeled knowledge in a Siamese graph neural network encoder through a well-designed contrastive loss. In the end, we train a decoder to reconstruct the original networks from the node representations learned by the encoder, and rank nodes according to its reconstruction error as the anomaly metric. Experiments on five real-world datasets demonstrate that the proposed framework outperforms the state-of-the-art anomaly detection algorithms.
more »
« less
Domain Adaptive 3D Pose Augmentation for In-the-wild Human Mesh Recovery
The ability to perceive 3D human bodies from a single image has a multitude of applications ranging from entertainment and robotics to neuroscience and healthcare. A fundamental challenge in human mesh recovery is in collecting the ground truth 3D mesh targets required for training, which requires burdensome motion capturing systems and is often limited to indoor laboratories. As a result, while progress is made on benchmark datasets collected in these restrictive settings, models fail to generalize to real-world "in-the-wild" scenarios due to distribution shifts. We propose Domain Adaptive 3D Pose Augmentation (DAPA), a data augmentation method that enhances the model's generalization ability in in-the-wild scenarios. DAPA combines the strength of methods based on synthetic datasets by getting direct supervision from the synthesized meshes, and domain adaptation methods by using ground truth 2D keypoints from the target dataset. We show quantitatively that finetuning with DAPA effectively improves results on benchmarks 3DPW and AGORA. We further demonstrate the utility of DAPA on a challenging dataset curated from videos of real-world parent-child interaction.
more »
« less
- Award ID(s):
- 2026498
- NSF-PAR ID:
- 10358710
- Date Published:
- Journal Name:
- International Conference on 3D Vision
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
S. Koyejo ; S. Mohamed ; A. Agarwal ; D. Belgrave ; K. Cho ; A. Oh (Ed.)Labeling articulated objects in unconstrained settings has a wide variety of applications including entertainment, neuroscience, psychology, ethology, and many fields of medicine. Large offline labeled datasets do not exist for all but the most common articulated object categories (e.g., humans). Hand labeling these landmarks within a video sequence is a laborious task. Learned landmark detectors can help, but can be error-prone when trained from only a few examples. Multi-camera systems that train fine-grained detectors have shown significant promise in detecting such errors, allowing for self-supervised solutions that only need a small percentage of the video sequence to be hand-labeled. The approach, however, is based on calibrated cameras and rigid geometry, making it expensive, difficult to manage, and impractical in real-world scenarios. In this paper, we address these bottlenecks by combining a non-rigid 3D neural prior with deep flow to obtain high-fidelity landmark estimates from videos with only two or three uncalibrated, handheld cameras. With just a few annotations (representing 1−2 % of the frames), we are able to produce 2D results comparable to state-of-the-art fully supervised methods, along with 3D reconstructions that are impossible with other existing approaches. Our Multi-view Bootstrapping in the Wild (MBW) approach demonstrates impressive results on standard human datasets, as well as tigers, cheetahs, fish, colobus monkeys, chimpanzees, and flamingos from videos captured casually in a zoo. We release the codebase for MBW as well as this challenging zoo dataset consisting of image frames of tail-end distribution categories with their corresponding 2D and 3D labels generated from minimal human intervention.more » « less
-
Labeling articulated objects in unconstrained settings have a wide variety of applications including entertainment, neuroscience, psychology, ethology, and many fields of medicine. Large offline labeled datasets do not exist for all but the most common articulated object categories (e.g., humans). Hand labeling these landmarks within a video sequence is a laborious task. Learned landmark detectors can help, but can be error-prone when trained from only a few examples. Multi-camera systems that train fine-grained detectors have shown significant promise in detecting such errors, allowing for self-supervised solutions that only need a small percentage of the video sequence to be hand-labeled. The approach, however, is based on calibrated cameras and rigid geometry, making it expensive, difficult to manage, and impractical in real-world scenarios. In this paper, we address these bottlenecks by combining a non-rigid 3D neural prior with deep flow to obtain high-fidelity landmark estimates from videos with only two or three uncalibrated, handheld cameras. With just a few annotations (representing 1-2% of the frames), we are able to produce 2D results comparable to state-of-the-art fully supervised methods, along with 3D reconstructions that are impossible with other existing approaches. Our Multi-view Bootstrapping in the Wild (MBW) approach demonstrates impressive results on standard human datasets, as well as tigers, cheetahs, fish, colobus monkeys, chimpanzees, and flamingos from videos captured casually in a zoo. We release the codebase for MBW as well as this challenging zoo dataset consisting image frames of tail-end distribution categories with their corresponding 2D, 3D labels generated from minimal human intervention.more » « less
-
Gonzalez, D. (Ed.)more » « less
Today’s research on human-robot teaming requires the ability to test artificial intelligence (AI) algorithms for perception and decision-making in complex real-world environments. Field experiments, also referred to as experiments “in the wild,” do not provide the level of detailed ground truth necessary for thorough performance comparisons and validation. Experiments on pre-recorded real-world data sets are also significantly limited in their usefulness because they do not allow researchers to test the effectiveness of active robot perception and control or decision strategies in the loop. Additionally, research on large human-robot teams requires tests and experiments that are too costly even for the industry and may result in considerable time losses when experiments go awry. The novel Real-Time Human Autonomous Systems Collaborations (RealTHASC) facility at Cornell University interfaces real and virtual robots and humans with photorealistic simulated environments by implementing new concepts for the seamless integration of wearable sensors, motion capture, physics-based simulations, robot hardware and virtual reality (VR). The result is an extended reality (XR) testbed by which real robots and humans in the laboratory are able to experience virtual worlds, inclusive of virtual agents, through real-time visual feedback and interaction. VR body tracking by DeepMotion is employed in conjunction with the OptiTrack motion capture system to transfer every human subject and robot in the real physical laboratory space into a synthetic virtual environment, thereby constructing corresponding human/robot avatars that not only mimic the behaviors of the real agents but also experience the virtual world through virtual sensors and transmit the sensor data back to the real human/robot agent, all in real time. New cross-domain synthetic environments are created in RealTHASC using Unreal Engine™, bridging the simulation-to-reality gap and allowing for the inclusion of underwater/ground/aerial autonomous vehicles, each equipped with a multi-modal sensor suite. The experimental capabilities offered by RealTHASC are demonstrated through three case studies showcasing mixed real/virtual human/robot interactions in diverse domains, leveraging and complementing the benefits of experimentation in simulation and in the real world.
-
Transparent objects are a very challenging problem in computer vision. They are hard to segment or classify due to their lack of precise boundaries, and there is limited data available for training deep neural networks. As such, current solutions for this problem employ rigid synthetic datasets, which lack flexibility and lead to severe performance degradation when deployed on real-world scenarios. In particular, these synthetic datasets omit features such as refraction, dispersion and caustics due to limitations in the rendering pipeline. To address this issue, we present SuperCaustics, a real-time, open-source simulation of transparent objects designed for deep learning applications. SuperCaustics features extensive modules for stochastic environment creation; uses hardware ray-tracing to support caustics, dispersion, and refraction; and enables generating massive datasets with multi-modal, pixel-perfect ground truth annotations. To validate our proposed system, we trained a deep neural network from scratch to segment transparent objects in difficult lighting scenarios. Our neural network achieved performance comparable to the state-of-the-art on a real-world dataset using only 10% of the training data and in a fraction of the training time. Further experiments show that a model trained with SuperCaustics can segment different types of caustics, even in images with multiple overlapping transparent objects. To the best of our knowledge, this is the first such result for a model trained on synthetic data. Both our open-source code and experimental data are freely available online.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.48550/arXiv.2206.10457
