More Like this

1. Multi-level 3D CNN for Learning Multi-scale Spatial Features

   Ghadai, Sambit; Yeow Lee, Xian; Balu, Aditya; Sarkar, Soumik; Krishnamurthy, Adarsh (January 2019, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops)

   3D object recognition accuracy can be improved by learning the multi-scale spatial features from 3D spatial geometric representations of objects such as point clouds, 3D models, surfaces, and RGB-D data. Current deep learning approaches learn such features either using structured data representations (voxel grids and octrees) or from unstructured representations (graphs and point clouds). Learning features from such structured representations is limited by the restriction on resolution and tree depth while unstructured representations creates a challenge due to non-uniformity among data samples. In this paper, we propose an end-to-end multi-level learning approach on a multi-level voxel grid to overcome these drawbacks. To demonstrate the utility of the proposed multi-level learning, we use a multi-level voxel representation of 3D objects to perform object recognition. The multi-level voxel representation consists of a coarse voxel grid that contains volumetric information of the 3D object. In addition, each voxel in the coarse grid that contains a portion of the object boundary is subdivided into multiple fine-level voxel grids. The performance of our multi-level learning algorithm for object recognition is comparable to dense voxel representations while using significantly lower memory. 

   more » « less
2. A Compact Handheld Sensor Package with Sensor Fusion for Comprehensive and Robust 3D Mapping

   https://doi.org/10.3390/s24082494  

   Wei, Peng; Fu, Kaiming; Villacres, Juan; Ke, Thomas; Krachenfels, Kay; Stofer, Curtis Ryan; Bayati, Nima; Gao, Qikai; Zhang, Bill; Vanacker, Eric; et al (April 2024, Sensors)

   This paper introduces an innovative approach to 3D environmental mapping through the integration of a compact, handheld sensor package with a two-stage sensor fusion pipeline. The sensor package, incorporating LiDAR, IMU, RGB, and thermal cameras, enables comprehensive and robust 3D mapping of various environments. By leveraging Simultaneous Localization and Mapping (SLAM) and thermal imaging, our solution offers good performance in conditions where global positioning is unavailable and in visually degraded environments. The sensor package runs a real-time LiDAR-Inertial SLAM algorithm, generating a dense point cloud map that accurately reconstructs the geometric features of the environment. Following the acquisition of that point cloud, we post-process these data by fusing them with images from the RGB and thermal cameras and produce a detailed, color-enriched 3D map that is useful and adaptable to different mission requirements. We demonstrated our system in a variety of scenarios, from indoor to outdoor conditions, and the results showcased the effectiveness and applicability of our sensor package and fusion pipeline. This system can be applied in a wide range of applications, ranging from autonomous navigation to smart agriculture, and has the potential to make a substantial benefit across diverse fields. 

   more » « less
3. Robust 3D Object Detection Using Probabilistic Point Clouds From Single-Photon Lidars

   https://doi.org/10.1109/ICCV51701.2025.02639  

   Goyal, Bhavya; Gutierrez-Barragan, Felipe; Lin, Wei; Velten, Andreas; Li, Yin; Gupta, Mohit (October 2025, IEEE)

   LiDAR-based 3D sensors provide point clouds, a canonical 3D representation used in various scene understanding tasks. Modern LiDARs face key challenges in several real-world scenarios, such as long-distance or low-albedo objects, producing sparse or erroneous point clouds. These errors, which are rooted in the noisy raw LiDAR measurements, get propagated to downstream perception models, resulting in potentially severe loss of accuracy. This is because conventional 3D processing pipelines do not retain any uncertainty information from the raw measurements when constructing point clouds. We propose Probabilistic Point Clouds (PPC), a novel 3D scene representation where each point is augmented with a probability attribute that encapsulates the measurement uncertainty (or confidence) in the raw data. We further introduce inference approaches that leverage PPC for robust 3D object detection; these methods are versatile and can be used as computationally lightweight drop-in modules in 3D inference pipelines. We demonstrate, via both simulations and real captures, that PPC-based 3D inference methods outperform several baselines using LiDAR as well as camera-LiDAR fusion models, across challenging indoor and outdoor scenarios involving small, distant, and low-albedo objects, as well as strong ambient light. Our project webpage is at https://bhavyagoyal.github.io/ppc/. 

   more » « less
4. Sensor Adversarial Traits: Analyzing Robustness of 3D Object Detection Sensor Fusion Models

   Park, Won; Liu, Nan; Chen, Qi Alfred; Mao, Z. Morley (January 2021, IEEE International Conference on Image Processing (ICIP))

   null (Ed.)

   A critical aspect of autonomous vehicles (AVs) is the object detection stage, which is increasingly being performed with sensor fusion models: multimodal 3D object detection models which utilize both 2D RGB image data and 3D data from a LIDAR sensor as inputs. In this work, we perform the first study to analyze the robustness of a high-performance, open source sensor fusion model architecture towards adversarial attacks and challenge the popular belief that the use of additional sensors automatically mitigate the risk of adversarial attacks. We find that despite the use of a LIDAR sensor, the model is vulnerable to our purposefully crafted image-based adversarial attacks including disappearance, universal patch, and spoofing. After identifying the underlying reason, we explore some potential defenses and provide some recommendations for improved sensor fusion models. 

   more » « less
5. Deep Depth Estimation from Visual-Inertial SLAM

   https://doi.org/10.1109/IROS45743.2020.9341448  

   Sartipi, Kourosh; Do, Tien; Ke, Tong; Vuong, Khiem; Roumeliotis, Stergios I. (October 2020, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS))

   null (Ed.)

   This paper addresses the problem of learning to complete a scene's depth from sparse depth points and images of indoor scenes. Specifically, we study the case in which the sparse depth is computed from a visual-inertial simultaneous localization and mapping (VI-SLAM) system. The resulting point cloud has low density, it is noisy, and has nonuniform spatial distribution, as compared to the input from active depth sensors, e.g., LiDAR or Kinect. Since the VI-SLAM produces point clouds only over textured areas, we compensate for the missing depth of the low-texture surfaces by leveraging their planar structures and their surface normals which is an important intermediate representation. The pre-trained surface normal network, however, suffers from large performance degradation when there is a significant difference in the viewing direction (especially the roll angle) of the test image as compared to the trained ones. To address this limitation, we use the available gravity estimate from the VI-SLAM to warp the input image to the orientation prevailing in the training dataset. This results in a significant performance gain for the surface normal estimate, and thus the dense depth estimates. Finally, we show that our method outperforms other state-of-the-art approaches both on training (ScanNet [1] and NYUv2 [2]) and testing (collected with Azure Kinect [3]) datasets. 

   more » « less