Skip to search formSkip to main contentSkip to account menu

KeypointNet: A Large-Scale 3D Keypoint Dataset Aggregated From Numerous Human Annotations

This work presents KeypointNet: the first large-scale and diverse 3D keypoint dataset that contains 83,231 keypoints and 8,329 3D models from 16 object categories, by leveraging numerous human annotations.
View on IEEE (opens in a new tab)

Pointwise Rotation-Invariant Network with Adaptive Sampling and 3D Spherical Voxel Convolution

A brand new point-set learning framework PRIN is proposed, namely, Pointwise Rotation-Invariant Network, focusing on rotation-invariant feature extraction in point clouds analysis, which shows that, on the dataset with randomly rotated point clouds, PRIN demonstrates better performance than state-of-the-art methods without any data augmentation.
View PDF on arXiv (opens in a new tab)

PRIN: Pointwise Rotation-Invariant Network

This paper proposes a new point-set learning framework named Pointwise Rotation-Invariant Network (PRIN), focusing on achieving rotation-invariance in point clouds, and shows performance better than state-of-the-art methods on part segmentation without data augmentation.

Canonical Voting: Towards Robust Oriented Bounding Box Detection in 3D Scenes

This work disentangle the direct offset into Local Canonical Coordinates (LCC), box scales and box orientations, which achieves state-of-the-art performance on three standard real-world benchmarks: ScanNet, SceneNN and SUN RGB-D.
View on IEEE (opens in a new tab)

PRIN/SPRIN: On Extracting Point-Wise Rotation Invariant Features

  • Yang YouYujing Lou Cewu Lu
  • Computer Science
    IEEE Transactions on Pattern Analysis and Machine…
  • 24 February 2021
This paper proposes a brand new point-set learning framework PRIN, namely, Point-wise Rotation Invariant Network, focusing on rotation invariant feature extraction in point clouds analysis, and extends PRIN to a sparse version called SPRIN, which directly operates on sparse point clouds.
View on IEEE (opens in a new tab)

Semantic Correspondence via 2D-3D-2D Cycle

This paper proposes a new method on predicting semantic correspondences by leveraging it to 3D domain and then project corresponding 3D models back to 2D domain, with their semantic labels.
View PDF on arXiv (opens in a new tab)

Fine-grained Object Semantic Understanding from Correspondences

This paper proposes a method that outputs dense semantic embeddings based on a novel geodesic consistency loss and shows that it could boost the fine-grained understanding of heterogeneous objects and the inference of dense semantic information is possible.
View PDF on arXiv (opens in a new tab)

Human Correspondence Consensus for 3D Object Semantic Understanding

This paper argues that by providing human labeled correspondences between different objects from the same category instead of explicit semantic labels, one can recover rich semantic information of an object and shows that CorresPondenceNet could not only boost fine-grained understanding of heterogeneous objects but also cross-object registration and partial object matching.
View via Publisher (opens in a new tab)

Understanding Pixel-Level 2D Image Semantics With 3D Keypoint Knowledge Engine

This paper proposes a new method on predicting image corresponding semantics in 3D domain and then projecting them back onto 2D images to achieve pixel-level understanding and shows that this method gives comparative and even superior results on standard semantic benchmarks.
View on IEEE (opens in a new tab)

Localization with Sampling-Argmax

Soft-argmax operation is commonly adopted in detection-based methods to localize the target position in a differentiable manner. However, training the neural network with soft-argmax makes the shape…
View PDF on arXiv (opens in a new tab)
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)