arXiv subjects

Shu Liu

Publications and source records attributed to Shu Liu.

188 records · Page 11Linked to original sources

arXiv2018

IPOD: Intensive Point-based Object Detector for Point Cloud

We present a novel 3D object detection framework, named IPOD, based on raw point cloud. It seeds object proposal for each point, which is the basic element. This paradigm provides us with high recall and high fidelity of information, leading to a suitable way to process point cloud data. We design an end-to-end trainable architecture, where features of all points within a proposal are extracted from the backbone network and achieve a proposal feature for final bounding inference. These features with both context information and precise point cloud coordinates yield improved performance. We conduct experiments on KITTI dataset, evaluating our performance in terms of 3D object detection, Bird's Eye View (BEV) detection and 2D object detection. Our method accomplishes new state-of-the-art , showing great advantage on the hard set.

cs.CV↗

arXiv2018

Sequential Context Encoding for Duplicate Removal

Duplicate removal is a critical step to accomplish a reasonable amount of predictions in prevalent proposal-based object detection frameworks. Albeit simple and effective, most previous algorithms utilize a greedy process without making sufficient use of properties of input data. In this work, we design a new two-stage framework to effectively select the appropriate proposal candidate for each object. The first stage suppresses most of easy negative object proposals, while the second stage selects true positives in the reduced proposal set. These two stages share the same network structure, \ie, an encoder and a decoder formed as recurrent neural networks (RNN) with global attention and context gate. The encoder scans proposal candidates in a sequential manner to capture the global context information, which is then fed to the decoder to extract optimal proposals. In our extensive experiments, the proposed method outperforms other alternatives by a large margin.

cs.CV↗

arXiv2018

Deep Anticipation: Light Weight Intelligent Mobile Sensing in IoT by Recurrent Architecture

The rapid growth of IoT era is shaping the future of mobile services. Advanced communication technology enables a heterogeneous connectivity where mobile devices broadcast information to everything. Mobile applications such as robotics and vehicles connecting to cloud and surroundings transfer the short-range on-board sensor perception system to long-range mobile-sensing perception system. However, the mobile sensing perception brings new challenges for how to efficiently analyze and intelligently interpret the deluge of IoT data in mission- critical services. In this article, we model the challenges as latency, packet loss and measurement noise which severely deteriorate the reliability and quality of IoT data. We integrate the artificial intelligence into IoT to tackle these challenges. We propose a novel architecture that leverages recurrent neural networks (RNN) and Kalman filtering to anticipate motions and interac- tions between objects. The basic idea is to learn environment dynamics by recurrent networks. To improve the robustness of IoT communication, we use the idea of Kalman filtering and deploy a prediction and correction step. In this way, the architecture learns to develop a biased belief between prediction and measurement in the different situation. We demonstrate our approach with synthetic and real-world datasets with noise that mimics the challenges of IoT communications. Our method brings a new level of IoT intelligence. It is also lightweight compared to other state-of-the-art convolutional recurrent architecture and is ideally suitable for the resource-limited mobile applications.

cs.CV↗

arXiv2018

Path Aggregation Network for Instance Segmentation

The way that information propagates in neural networks is of great importance. In this paper, we propose Path Aggregation Network (PANet) aiming at boosting information flow in proposal-based instance segmentation framework. Specifically, we enhance the entire feature hierarchy with accurate localization signals in lower layers by bottom-up path augmentation, which shortens the information path between lower layers and topmost feature. We present adaptive feature pooling, which links feature grid and all feature levels to make useful information in each feature level propagate directly to following proposal subnetworks. A complementary branch capturing different views for each proposal is created to further improve mask prediction. These improvements are simple to implement, with subtle extra computational overhead. Our PANet reaches the 1st place in the COCO 2017 Challenge Instance Segmentation task and the 2nd place in Object Detection task without large-batch training. It is also state-of-the-art on MVD and Cityscapes. Code is available at https://github.com/ShuLiu1993/PANet

cs.CV↗

arXiv2018

Evaluating Semantic Rationality of a Sentence: A Sememe-Word-Matching Neural Network based on HowNet

Automatic evaluation of semantic rationality is an important yet challenging task, and current automatic techniques cannot well identify whether a sentence is semantically rational. The methods based on the language model do not measure the sentence by rationality but by commonness. The methods based on the similarity with human written sentences will fail if human-written references are not available. In this paper, we propose a novel model called Sememe-Word-Matching Neural Network (SWM-NN) to tackle semantic rationality evaluation by taking advantage of sememe knowledge base HowNet. The advantage is that our model can utilize a proper combination of sememes to represent the fine-grained semantic meanings of a word within the specific contexts. We use the fine-grained semantic representation to help the model learn the semantic dependency among words. To evaluate the effectiveness of the proposed model, we build a large-scale rationality evaluation dataset. Experimental results on this dataset show that the proposed model outperforms the competitive baselines with a 5.4\% improvement in accuracy.

cs.CL↗

arXiv2018

List Decodability of Symbol-Pair Codes

We investigate the list decodability of symbol-pair codes in the present paper. Firstly, we show that list decodability of every symbol-pair code does not exceed the Gilbert-Varshamov bound. On the other hand, we are able to prove that with high probability, a random symbol-pair code can be list decoded up to the Gilbert-Varshamov bound. Our second result of this paper is to derive the Johnson-type bound, i.e., a lower bound on list decoding radius in terms of minimum distance. Finally, we present a list decoding algorithm of Reed-Solomon codes beyond the Johnson-type bound.

cs.IT↗

arXiv2018

On the List Decodability of Self-orthogonal Rank Metric Codes

V. Guruswami and N. Resch prove that the list decodability of $\mathbb{F}_q$-linear rank metric codes is as good as that of random rank metric codes in~\cite{venkat2017}. Due to the potential applications of self-orthogonal rank metric codes, we focus on list decoding of them. In this paper, we prove that with high probability, an $\F_q$-linear self-orthogonal rank metric code over $\mathbb{F}_q^{n\times m}$ of rate $R=(1-τ)(1-\frac{n}{m}τ)-ε$ is shown to be list decodable up to fractional radius $τ\in(0,1)$ and small $ε\in(0,1)$ with list size depending on $τ$ and $q$ at most $O_{τ, q}(\frac{1}ε)$. In addition, we show that an $\mathbb{F}_{q^m}$-linear self-orthogonal rank metric code of rate up to the Gilbert-Varshamov bound is $(τn, \exp(O_{τ, q}(\frac{1}ε)))$-list decodable.

cs.IT↗

arXiv2016

Label distribution based facial attractiveness computation by deep residual learning

Two challenges lie in the facial attractiveness computation research: the lack of true attractiveness labels (scores), and the lack of an accurate face representation. In order to address the first challenge, this paper recasts facial attractiveness computation as a label distribution learning (LDL) problem rather than a traditional single-label supervised learning task. In this way, the negative influence of the label incomplete problem can be reduced. Inspired by the recent promising work in face recognition using deep neural networks to learn effective features, the second challenge is expected to be solved from a deep learning point of view. A very deep residual network is utilized to enable automatic learning of hierarchical aesthetics representation. Integrating these two ideas, an end-to-end deep learning framework is established. Our approach achieves the best results on a standard benchmark SCUT-FBP dataset compared with other state-of-the-art work.

cs.CV↗