Author name cluster

Yanan Liu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

4 papers

2 author rows

ICRA Conference 2021 Conference Paper

Weighted Node Mapping and Localisation on a Pixel Processor Array

Hector Castillo-Elizalde
Yanan Liu
Laurie Bose
Walterio W. Mayol-Cuevas

This paper implements and demonstrates visual route mapping and localisation upon a Pixel Processor Array (PPA). The PPA sensor comprises of an array of Processing Elements (PEs), each of which can capture and process visual information directly. This provides significant parallel processing power allowing novel ways in which information can be processed on-sensor. Our method predicts the correct node within a topological map generated from an image sequence by measuring image similarities, spatial coherence, and exploiting the parallel nature of the PPA. Our implementation runs at +300Hz on large public datasets with +2K locations requiring 2. 5W at 500 GOPS/W. We compare vs traditionally implemented methods demonstrating better F-1 performance even on simulation. As far as we are aware, we present the first on-sensor mapping and localisation system running entirely on-sensor.

Details

TIST Journal 2020 Journal Article

A Traffic Density Estimation Model Based on Crowdsourcing Privacy Protection

Yapei Huang
Yun Tian
Zhijie Liu
Xiaowei Jin
Yanan Liu
Shifeng Zhao
Daxin Tian

Acquiring traffic condition information is of great significance in transportation guidance, urban planning, and route recommendation. To date, traffic density data are generally acquired by road sound analysis, video data analysis, or in-vehicle network communication, which are usually financially or temporally expensive. Another way to get traffic conditions is to collect track data by crowdsourcing. However, this way lead to a greater risk of leaking users’ privacy. To avoid the risk, this article proposes a traffic density estimation model based on crowdsourcing privacy protection. First, in the acquisition process of the track data by crowdsourcing, dual servers are employed for transmission, and homomorphic encryption is carried out to encrypt the data to protect the data from being leaked during transmission. Second, sampling is implemented for randomization and anonymization to reduce the spatial continuity and temporal continuity of position data. In this way, the intermediate server cannot acquire users’ original data, and the main server cannot obtain users’ personal information. Finally, before data transmission, Laplace noising is performed on the users’ local position data to further protect the original location information. The proposed algorithm in this study realizes that only users have their original track data, and the servers involved in the work cannot infer the original track data, which ensures the real security of user privacy. The proposed algorithm was verified with the track data from the Didi Gaia Data Opening Plan. The experimental results showed that the proposed algorithm could still maintain the validity of data analysis results and the security of user data privacy after homomorphic encryption, noise addition, and sample collection, and displayed good robustness and scalability.

Details DOI

ICRA Conference 2020 Conference Paper

Proximity Estimation Using Vision Features Computed On Sensor

Jianing Chen 0005
Yanan Liu
Stephen J. Carey
Piotr Dudek

This paper presents a monocular vision based proximity estimation system using abstract features, such as corner points, blobs and edges, as inputs to a neural network. An experimental vehicle was built using a vision system integrating the SCAMP-5 vision chip, a micro-controller, and an RC model car. The vision chip includes image sensor with embedded 256×256 processor SIMD array. The pixel processor array chip was programmed to capture images and run the feature algorithms directly on the focal plane, and then digest them so that only sparse feature description data were read-out in the form of 40 values. By logging the vision output and the output from three infrared proximity sensors, training data were obtained to train three fully connected layer-recurrent neural networks with fewer than 700 parameters each. The trained neural network was able to estimate the proximity to the level of accuracy sufficient for a reactive collision avoidance behaviour to be achieved. The latency of the control system, from image capture to neural network output, was under 4ms, enabling the vehicles to avoid obstacles while moving at 0. 64m/s to 1. 8m/s in the experiment.

Details

AAAI Conference 2020 Conference Paper

Target-Aspect-Sentiment Joint Detection for Aspect-Based Sentiment Analysis

Hai Wan
Yufei Yang
Jianfeng Du
Yanan Liu
Kunxun Qi
Jeff Z. Pan

Aspect-based sentiment analysis (ABSA) aims to detect the targets (which are composed by continuous words), aspects and sentiment polarities in text. Published datasets from SemEval-2015 and SemEval-2016 reveal that a sentiment polarity depends on both the target and the aspect. However, most of the existing methods consider predicting sentiment polarities from either targets or aspects but not from both, thus they easily make wrong predictions on sentiment polarities. In particular, where the target is implicit, i. e. , it does not appear in the given text, the methods predicting sentiment polarities from targets do not work. To tackle these limitations in ABSA, this paper proposes a novel method for target-aspectsentiment joint detection. It relies on a pre-trained language model and can capture the dependence on both targets and aspects for sentiment prediction. Experimental results on the SemEval-2015 and SemEval-2016 restaurant datasets show that the proposed method achieves a high performance in detecting target-aspect-sentiment triples even for the implicit target cases; moreover, it even outperforms the state-of-theart methods for those subtasks of target-aspect-sentiment detection that they are competent to.

PDF Details