Search Results for author: Jun Shimamura

Found 5 papers, 1 papers with code

Non-learning Stereo-aided Depth Completion under Mis-projection via Selective Stereo Matching

no code implementations4 Oct 2022 Yasuhiro Yao, Ryoichi Ishikawa, Shingo Ando, Kana Kurata, Naoki Ito, Jun Shimamura, Takeshi Oishi

Moreover, under various LiDAR-camera calibration errors, the proposed method reduced the depth estimation MAE to 0. 34-0. 93 times from previous depth completion methods.

Camera Calibration Depth Completion +2

Discontinuous and Smooth Depth Completion with Binary Anisotropic Diffusion Tensor

no code implementations25 Jun 2020 Yasuhiro Yao, Menandro Roxas, Ryoichi Ishikawa, Shingo Ando, Jun Shimamura, Takeshi Oishi

Our experiments show that our method can outperform previous unsupervised and semi-supervised depth completion methods in terms of accuracy.

Depth Completion

Concatenated Feature Pyramid Network for Instance Segmentation

no code implementations16 Mar 2019 Yongqing Sun, Pranav Shenoy K P, Jun Shimamura, Atsushi Sagata

Low level features like edges and textures play an important role in accurately localizing instances in neural networks.

Instance Segmentation object-detection +3

Weakly Supervised Instance Segmentation Using Hybrid Network

no code implementations12 Dec 2018 Shisha Liao, Yongqing Sun, Chenqiang Gao, Pranav Shenoy K P, Song Mu, Jun Shimamura, Atsushi Sagata

Weakly-supervised instance segmentation, which could greatly save labor and time cost of pixel mask annotation, has attracted increasing attention in recent years.

Image Segmentation Instance Segmentation +4

Cannot find the paper you are looking for? You can Submit a new open access paper.