Sliding Shapes for 3D Object Detection in Depth Images

By Shuran Song, Jianxiong Xiao. 2014

424 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Control and Systems Engineering, 3D Vision & Pose

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

In 3D Vision & Pose

Citations in the corpus

VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (2017, background): “Other approaches rasterize point clouds into a 3D voxel grid and encode each voxel with handcrafted features [43, 9, 39, 40, 21, 5].”
ShapeNet: An Information-Rich 3D Model Repository (2015, background): “Scene understanding from 2D images is a grand challenge in vision that has recently benefited tremendously from 3D CAD models [28, 34].”
3D fully convolutional network for vehicle detection in point cloud (2016, methodology): “Candidates can be proposed by delicate segmentation algorithms [2], [5], [11], [14], [21], [22], [28], [29], [31], sliding window [30], random sampling [13], or the recently popular Region Proposal Network (RPN) [26].”
Multi-view 3D Object Detection Network for Autonomous Driving (2016, methodology): “Deep Sliding Shapes [19] exploits more powerful deep learning features.”
Vehicle Detection from 3D Lidar Using Fully Convolutional Network (2016, methodology): “These classiﬁers can be integrated with region proposal method like sliding window [27] for detection tasks.”
Deep Sliding Shapes for Amodal 3D Object Detection in RGB-D Images (2015, methodology): “Table 3 shows the comparison with the two state-of-the-arts for amodal 3D detection: 3D Sliding Shapes [25] with hand-crafted features, and 2D Depth-RCNN [10] with ConvNets features.”
Voting for Voting in Online Point Cloud Object Detection (2015, positive): “Most closely related to our work is the sliding window approach to 3D object detection recently proposed by Song and Xiao [16], who share our aspiration of passing a 3D window through a voxel grid.”
OctNetFusion: Learning Depth Fusion from Data (2017, background): “This has led to the creation of large databases of real-world 3D content [7, 10, 42, 43], enabling progress in a variety of areas including 3D modeling [28, 54], 3D reconstruction [6, 52], 3D recognition [18, 44] and 3D scene understanding [16, 42].”
A Survey of Autonomous Driving: Common Practices and Emerging Technologies (2019, methodology): “There are two main representations of point cloud data in 2D, the ﬁrst being a so-called depth image shown in Figure 14, largely inspired by camera-based methods that perform 3D object detection through depth estimation [165] and methods that operate on RGB-D data [166].”
Semantic Scene Completion from a Single Depth Image (2016, background): “One possible approach to obtain the complete geometry and semantic labels for a scene is to retrieve and fit instance-level 3D mesh models to the observed depth map [7, 30, 4, 16, 23, 17, 14].”
Matterport3D: Learning from RGB-D Data in Indoor Environments (2017, methodology): “These previous RGB-D datasets have been used to train models for several standard scene understanding tasks, including semantic segmentation [27, 33, 13, 40, 7], 3D object detection [31, 24, 14, 36, 37], normal estimation [41, 23, 9, 3], camera relocalization [30, 39], and others [47, 11, 12].”
Marr Revisited: 2D-3D Alignment via Surface Normal Prediction (2016, methodology): “Kinect) include exemplar detection by rendering depth from CAD and sliding in 3D [42], 3D model retrieval via exemplar regions matched to object proposals (while optimizing over room layout) [14], and training CNNs to predict pose for CAD model alignment [15] and to predict object class, location, a”
Semantically-enriched 3D models for common-sense knowledge (2015, methodology): “We use a 3D model size estimation method designed to propagate physical size priors between models in 3D scenes [20] to obtain physical sizes for our corpus.”
PIXOR: Real-time 3D Object Detection from Point Clouds (2018, background): “With the popularity of inexpensive RGB-D sensors such as Microsoft Kinect, Intel RealSense and Apple PrimeSense, several approaches that utilize depth information and fuse them with RGB images have been developed [32, 33].”
Frustum PointNets for 3D Object Detection from RGB-D Data (2017, methodology): “3D based methods: [31, 28] train 3D object classifiers by SVMs on hand-designed geometry features extracted from point cloud and then localize objects using slidingwindow search.”
Structured Prediction of Unobserved Voxels from a Single Depth Image (2016, methodology): “If an exact model of the object of interest is not present in the database, it is possible to fit objects of the same class [54, 21].”
Aligning 3D models to RGB-D images of cluttered scenes (2015, methodology): “When we use this 3D bounding box for 3D detection we observe an improvement over the current state-of-the-art method (‘Sliding Shapes’) [34] of 19% absolute AP points (48% relative), while being at least an order of magnitude faster.”
3D all the way: Semantic segmentation of urban scenes from start to end in 3D (2015, background): “Since then, recent works show the advantage of combining 3D and 2D for classification [42, 21, 19, 38, 56] or place recognition [37] using LIDAR, 3D CAD or Kinect models.”