activity
20172019
most citedFully Using Classifiers for Weakly Supervised Semantic Segmentation with Modified Cues

3 citations · 3 across the 2 of their papers we have counts for

collaborators

6 papers

cs.RO2019

Movable-Object-Aware Visual SLAM via Weakly Supervised Semantic Segmentation

Ting Sun, Yuxiang Sun, Ming Liu +1

Moving objects can greatly jeopardize the performance of a visual simultaneous localization and mapping (vSLAM) system which relies on the static-world assumption. Motion removal h…

cs.CV20193 cited

Fully Using Classifiers for Weakly Supervised Semantic Segmentation with Modified Cues

Ting Sun, Lei Tai, Zhihan Gao +2

This paper proposes a novel weakly-supervised semantic segmentation method using image-level label only. The class-specific activation maps from the well-trained classifiers are us…

cs.CV2018

Semi-Semantic Line-Cluster Assisted Monocular SLAM for Indoor Environments

Ting Sun, Dezhen Song, Dit-Yan Yeung +1

This paper presents a novel method to reduce the scale drift for indoor monocular simultaneous localization and mapping (SLAM). We leverage the prior knowledge that in the indoor e…

cs.CV2018

Point-cloud-based place recognition using CNN feature extraction

Ting Sun, Ming Liu, Haoyang Ye +1

This paper proposes a novel point-cloud-based place recognition system that adopts a deep learning approach for feature extraction. By using a convolutional neural network pre-trai…

cs.HC2018

Gesture-based Piloting of an Aerial Robot using Monocular Vision

Ting Sun, Shengyi Nie, Dit-Yan Yeung +1

Aerial robots are becoming popular among general public, and with the development of artificial intelligence (AI), there is a trend to equip aerial robots with a natural user inter…

cs.CV2017

Fine-Grained Categorization via CNN-Based Automatic Extraction and Integration of Object-Level and Part-Level Features

Ting Sun, Lin Sun, Dit-Yan Yeung

Fine-grained categorization can benefit from part-based features which reveal subtle visual differences between object categories. Handcrafted features have been widely used for pa…