January 24, 2020

2948 words 14 mins read

Paper Group NANR 208

Dynamic Points Agglomeration for Hierarchical Point Sets Learning. Occupancy Flow: 4D Reconstruction by Learning Particle Dynamics. Resample-smoothing of Voronoi intensity estimators. Pose-Guided Feature Alignment for Occluded Person Re-Identification. Looking for ELMo’s friends: Sentence-Level Pretraining Beyond Language Modeling. Explicit Plannin …

Dynamic Points Agglomeration for Hierarchical Point Sets Learning


Title	Dynamic Points Agglomeration for Hierarchical Point Sets Learning
Authors	Jinxian Liu, Bingbing Ni, Caiyuan Li, Jiancheng Yang, Qi Tian
Abstract	Many previous works on point sets learning achieve excellent performance with hierarchical architecture. Their strategies towards points agglomeration, however, only perform points sampling and grouping in original Euclidean space in a fixed way. These heuristic and task-irrelevant strategies severely limit their ability to adapt to more varied scenarios. To this end, we develop a novel hierarchical point sets learning architecture, with dynamic points agglomeration. By exploiting the relation of points in semantic space, a module based on graph convolution network is designed to learn a soft points cluster agglomeration. We construct a hierarchical architecture that gradually agglomerates points by stacking this learnable and lightweight module. In contrast to fixed points agglomeration strategy, our method can handle more diverse situations robustly and efficiently. Moreover, we propose a parameter sharing scheme for reducing memory usage and computational burden induced by the agglomeration module. Extensive experimental results on several point cloud analytic tasks, including classification and segmentation, well demonstrate the superior performance of our dynamic hierarchical learning framework over current state-of-the-art methods.
Tasks
Published	2019-10-01
URL	http://openaccess.thecvf.com/content_ICCV_2019/html/Liu_Dynamic_Points_Agglomeration_for_Hierarchical_Point_Sets_Learning_ICCV_2019_paper.html
PDF	http://openaccess.thecvf.com/content_ICCV_2019/papers/Liu_Dynamic_Points_Agglomeration_for_Hierarchical_Point_Sets_Learning_ICCV_2019_paper.pdf
PWC	https://paperswithcode.com/paper/dynamic-points-agglomeration-for-hierarchical
Repo
Framework

Occupancy Flow: 4D Reconstruction by Learning Particle Dynamics


Title	Occupancy Flow: 4D Reconstruction by Learning Particle Dynamics
Authors	Michael Niemeyer, Lars Mescheder, Michael Oechsle, Andreas Geiger
Abstract	Deep learning based 3D reconstruction techniques have recently achieved impressive results. However, while state-of-the-art methods are able to output complex 3D geometry, it is not clear how to extend these results to time-varying topologies. Approaches treating each time step individually lack continuity and exhibit slow inference, while traditional 4D reconstruction methods often utilize a template model or discretize the 4D space at fixed resolution. In this work, we present Occupancy Flow, a novel spatio-temporal representation of time-varying 3D geometry with implicit correspondences. Towards this goal, we learn a temporally and spatially continuous vector field which assigns a motion vector to every point in space and time. In order to perform dense 4D reconstruction from images or sparse point clouds, we combine our method with a continuous 3D representation. Implicitly, our model yields correspondences over time, thus enabling fast inference while providing a sound physical description of the temporal dynamics. We show that our method can be used for interpolation and reconstruction tasks, and demonstrate the accuracy of the learned correspondences. We believe that Occupancy Flow is a promising new 4D representation which will be useful for a variety of spatio-temporal reconstruction tasks.
Tasks	3D Reconstruction
Published	2019-10-01
URL	http://openaccess.thecvf.com/content_ICCV_2019/html/Niemeyer_Occupancy_Flow_4D_Reconstruction_by_Learning_Particle_Dynamics_ICCV_2019_paper.html
PDF	http://openaccess.thecvf.com/content_ICCV_2019/papers/Niemeyer_Occupancy_Flow_4D_Reconstruction_by_Learning_Particle_Dynamics_ICCV_2019_paper.pdf
PWC	https://paperswithcode.com/paper/occupancy-flow-4d-reconstruction-by-learning
Repo
Framework

Resample-smoothing of Voronoi intensity estimators


Title	Resample-smoothing of Voronoi intensity estimators
Authors	M. Mehdi Moradi, Ottmar Cronie, Ege Rubak, Raphael Lachieze-Rey, Jorge Mateu, Adrian Baddeley.
Abstract	Voronoi estimators are non-parametric and adaptive estimators of the intensity of a point process. The intensity estimate at a given location is equal to the reciprocal of the size of the Voronoi/Dirichlet cell containing that location. Their major drawback is that they tend to paradoxically under-smooth the data in regions where the point density of the observed point pattern is high, and over-smooth where the point density is low. To remedy this behaviour, we propose to apply an additional smoothing operation to the Voronoi estimator, based on resampling the point pattern by independent random thinning. Through a simulation study we show that our resample-smoothing technique improves the estimation substantially. In addition, we study statistical properties such as unbiasedness and variance, and propose a rule-of-thumb and a data-driven cross-validation approach to choose the amount of smoothing to apply. Finally we apply our proposed intensity estimation scheme to two datasets: locations of pine saplings (planar point pattern) and motor vehicle traffic accidents (linear network point pattern). The code implementation can be found in the R package spatstat using the function “adaptive.density”.
Tasks
Published	2019-01-19
URL	https://link.springer.com/article/10.1007%2Fs11222-018-09850-0
PDF	https://link.springer.com/content/pdf/10.1007%2Fs11222-018-09850-0.pdf
PWC	https://paperswithcode.com/paper/resample-smoothing-of-voronoi-intensity
Repo
Framework

Pose-Guided Feature Alignment for Occluded Person Re-Identification


Title	Pose-Guided Feature Alignment for Occluded Person Re-Identification
Authors	Jiaxu Miao, Yu Wu, Ping Liu, Yuhang Ding, Yi Yang
Abstract	Persons are often occluded by various obstacles in person retrieval scenarios. Previous person re-identification (re-id) methods, either overlook this issue or resolve it based on an extreme assumption. To alleviate the occlusion problem, we propose to detect the occluded regions, and explicitly exclude those regions during feature generation and matching. In this paper, we introduce a novel method named Pose-Guided Feature Alignment (PGFA), exploiting pose landmarks to disentangle the useful information from the occlusion noise. During the feature constructing stage, our method utilizes human landmarks to generate attention maps. The generated attention maps indicate if a specific body part is occluded and guide our model to attend to the non-occluded regions. During matching, we explicitly partition the global feature into parts and use the pose landmarks to indicate which partial features belonging to the target person. Only the visible regions are utilized for the retrieval. Besides, we construct a large-scale dataset for the Occluded Person Re-ID problem, namely Occluded-DukeMTMC, which is by far the largest dataset for the Occlusion Person Re-ID. Extensive experiments are conducted on our constructed occluded re-id dataset, two partial re-id datasets, and two commonly used holistic re-id datasets. Our method largely outperforms existing person re-id methods on three occlusion datasets, while remains top performance on two holistic datasets.
Tasks	Person Re-Identification, Person Retrieval
Published	2019-10-01
URL	http://openaccess.thecvf.com/content_ICCV_2019/html/Miao_Pose-Guided_Feature_Alignment_for_Occluded_Person_Re-Identification_ICCV_2019_paper.html
PDF	http://openaccess.thecvf.com/content_ICCV_2019/papers/Miao_Pose-Guided_Feature_Alignment_for_Occluded_Person_Re-Identification_ICCV_2019_paper.pdf
PWC	https://paperswithcode.com/paper/pose-guided-feature-alignment-for-occluded
Repo
Framework

Looking for ELMo’s friends: Sentence-Level Pretraining Beyond Language Modeling


Title	Looking for ELMo’s friends: Sentence-Level Pretraining Beyond Language Modeling
Authors	Samuel R. Bowman, Ellie Pavlick, Edouard Grave, Benjamin Van Durme, Alex Wang, Jan Hula, Patrick Xia, Raghavendra Pappagari, R. Thomas McCoy, Roma Patel, Najoung Kim, Ian Tenney, Yinghui Huang, Katherin Yu, Shuning Jin, Berlin Chen
Abstract	Work on the problem of contextualized word representation—the development of reusable neural network components for sentence understanding—has recently seen a surge of progress centered on the unsupervised pretraining task of language modeling with methods like ELMo (Peters et al., 2018). This paper contributes the first large-scale systematic study comparing different pretraining tasks in this context, both as complements to language modeling and as potential alternatives. The primary results of the study support the use of language modeling as a pretraining task and set a new state of the art among comparable models using multitask learning with language models. However, a closer look at these results reveals worryingly strong baselines and strikingly varied results across target tasks, suggesting that the widely-used paradigm of pretraining and freezing sentence encoders may not be an ideal platform for further work.
Tasks	Language Modelling
Published	2019-05-01
URL	https://openreview.net/forum?id=Bkl87h09FX
PDF	https://openreview.net/pdf?id=Bkl87h09FX
PWC	https://paperswithcode.com/paper/looking-for-elmos-friends-sentence-level-1
Repo
Framework

Explicit Planning for Efficient Exploration in Reinforcement Learning


Title	Explicit Planning for Efficient Exploration in Reinforcement Learning
Authors	Liangpeng Zhang, Ke Tang, Xin Yao
Abstract	Efficient exploration is crucial to achieving good performance in reinforcement learning. Existing systematic exploration strategies (R-MAX, MBIE, UCRL, etc.), despite being promising theoretically, are essentially greedy strategies that follow some predefined heuristics. When the heuristics do not match the dynamics of Markov decision processes (MDPs) well, an excessive amount of time can be wasted in travelling through already-explored states, lowering the overall efficiency. We argue that explicit planning for exploration can help alleviate such a problem, and propose a Value Iteration for Exploration Cost (VIEC) algorithm which computes the optimal exploration scheme by solving an augmented MDP. We then present a detailed analysis of the exploration behaviour of some popular strategies, showing how these strategies can fail and spend O(n^2 md) or O(n^2 m + nmd) steps to collect sufficient data in some tower-shaped MDPs, while the optimal exploration scheme, which can be obtained by VIEC, only needs O(nmd), where n, m are the numbers of states and actions and d is the data demand. The analysis not only points out the weakness of existing heuristic-based strategies, but also suggests a remarkable potential in explicit planning for exploration.
Tasks	Efficient Exploration
Published	2019-12-01
URL	http://papers.nips.cc/paper/8967-explicit-planning-for-efficient-exploration-in-reinforcement-learning
PDF	http://papers.nips.cc/paper/8967-explicit-planning-for-efficient-exploration-in-reinforcement-learning.pdf
PWC	https://paperswithcode.com/paper/explicit-planning-for-efficient-exploration
Repo
Framework

The Extent of Repetition in Contract Language


Title	The Extent of Repetition in Contract Language
Authors	Dan Simonson, Daniel Broderick, Jonathan Herr
Abstract	Contract language is repetitive (Anderson and Manns, 2017), but so is all language (Zipf, 1949). In this paper, we measure the extent to which contract language in English is repetitive compared with the language of other English language corpora. Contracts have much smaller vocabulary sizes compared with similarly sized non-contract corpora across multiple contract types, contain 1/5th as many hapax legomena, pattern differently on a log-log plot, use fewer pronouns, and contain sentences that are about 20{%} more similar to one another than in other corpora. These suggest that the study of contracts in natural language processing controls for some linguistic phenomena and allows for more in depth study of others.
Tasks
Published	2019-06-01
URL	https://www.aclweb.org/anthology/W19-2203/
PDF	https://www.aclweb.org/anthology/W19-2203
PWC	https://paperswithcode.com/paper/the-extent-of-repetition-in-contract-language
Repo
Framework

Unified Visual-Semantic Embeddings: Bridging Vision and Language With Structured Meaning Representations


Title	Unified Visual-Semantic Embeddings: Bridging Vision and Language With Structured Meaning Representations
Authors	Hao Wu, Jiayuan Mao, Yufeng Zhang, Yuning Jiang, Lei Li, Weiwei Sun, Wei-Ying Ma
Abstract	We propose the Unified Visual-Semantic Embeddings (Unified VSE) for learning a joint space of visual representation and textual semantics. The model unifies the embeddings of concepts at different levels: objects, attributes, relations, and full scenes. We view the sentential semantics as a combination of different semantic components such as objects and relations; their embeddings are aligned with different image regions. A contrastive learning approach is proposed for the effective learning of this fine-grained alignment from only image-caption pairs. We also present a simple yet effective approach that enforces the coverage of caption embeddings on the semantic components that appear in the sentence. We demonstrate that the Unified VSE outperforms baselines on cross-modal retrieval tasks; the enforcement of the semantic coverage improves the model’s robustness in defending text-domain adversarial attacks. Moreover, our model empowers the use of visual cues to accurately resolve word dependencies in novel sentences.
Tasks	Cross-Modal Retrieval
Published	2019-06-01
URL	http://openaccess.thecvf.com/content_CVPR_2019/html/Wu_Unified_Visual-Semantic_Embeddings_Bridging_Vision_and_Language_With_Structured_Meaning_CVPR_2019_paper.html
PDF	http://openaccess.thecvf.com/content_CVPR_2019/papers/Wu_Unified_Visual-Semantic_Embeddings_Bridging_Vision_and_Language_With_Structured_Meaning_CVPR_2019_paper.pdf
PWC	https://paperswithcode.com/paper/unified-visual-semantic-embeddings-bridging-1
Repo
Framework

Veritatem Dies Aperit - Temporally Consistent Depth Prediction Enabled by a Multi-Task Geometric and Semantic Scene Understanding Approach


Title	Veritatem Dies Aperit - Temporally Consistent Depth Prediction Enabled by a Multi-Task Geometric and Semantic Scene Understanding Approach
Authors	Amir Atapour-Abarghouei, Toby P. Breckon
Abstract	Robust geometric and semantic scene understanding is ever more important in many real-world applications such as autonomous driving and robotic navigation. In this paper, we propose a multi-task learning-based approach capable of jointly performing geometric and semantic scene understanding, namely depth prediction (monocular depth estimation and depth completion) and semantic scene segmentation. Within a single temporally constrained recurrent network, our approach uniquely takes advantage of a complex series of skip connections, adversarial training and the temporal constraint of sequential frame recurrence to produce consistent depth and semantic class labels simultaneously. Extensive experimental evaluation demonstrates the efficacy of our approach compared to other contemporary state-of-the-art techniques.
Tasks	Autonomous Driving, Depth Completion, Depth Estimation, Monocular Depth Estimation, Multi-Task Learning, Scene Segmentation, Scene Understanding
Published	2019-06-01
URL	http://openaccess.thecvf.com/content_CVPR_2019/html/Atapour-Abarghouei_Veritatem_Dies_Aperit_-_Temporally_Consistent_Depth_Prediction_Enabled_by_CVPR_2019_paper.html
PDF	http://openaccess.thecvf.com/content_CVPR_2019/papers/Atapour-Abarghouei_Veritatem_Dies_Aperit_-_Temporally_Consistent_Depth_Prediction_Enabled_by_CVPR_2019_paper.pdf
PWC	https://paperswithcode.com/paper/veritatem-dies-aperit-temporally-consistent-1
Repo
Framework

Frame Identification as Categorization: Exemplars vs Prototypes in Embeddingland


Title	Frame Identification as Categorization: Exemplars vs Prototypes in Embeddingland
Authors	Jennifer Sikos, Sebastian Pad{'o}
Abstract	Categorization is a central capability of human cognition, and a number of theories have been developed to account for properties of categorization. Even though many tasks in semantics also involve categorization of some kind, theories of categorization do not play a major role in contemporary research in computational linguistics. This paper follows the idea that embedding-based models of semantics lend themselves well to being formulated in terms of classical categorization theories. The benefit is a space of model families that enables (a) the formulation of hypotheses about the impact of major design decisions, and (b) a transparent assessment of these decisions. We instantiate this idea on the task of frame-semantic frame identification. We define four models that cross two design variables: (a) the choice of prototype vs. exemplar categorization, corresponding to different degrees of generalization applied to the input; and (b) the presence vs. absence of a fine-tuning step, corresponding to generic vs. task-adaptive categorization. We find that for frame identification, generalization and task-adaptive categorization both yield substantial benefits. Our prototype-based, fine-tuned model, which combines the best choices for these variables, establishes a new state of the art in frame identification.
Tasks
Published	2019-05-01
URL	https://www.aclweb.org/anthology/W19-0425/
PDF	https://www.aclweb.org/anthology/W19-0425
PWC	https://paperswithcode.com/paper/frame-identification-as-categorization
Repo
Framework

HIGHLY EFFICIENT 8-BIT LOW PRECISION INFERENCE OF CONVOLUTIONAL NEURAL NETWORKS


Title	HIGHLY EFFICIENT 8-BIT LOW PRECISION INFERENCE OF CONVOLUTIONAL NEURAL NETWORKS
Authors	Haihao Shen, Jiong Gong, Xiaoli Liu, Guoming Zhang, Ge Jin, and Eric Lin
Abstract	High throughput and low latency inference of deep neural networks are critical for the deployment of deep learning applications. This paper presents a general technique toward 8-bit low precision inference of convolutional neural networks, including 1) channel-wise scale factors of weights, especially for depthwise convolution, 2) Winograd convolution, and 3) topology-wise 8-bit support. We experiment the techniques on top of a widely-used deep learning framework. The 8-bit optimized model is automatically generated with a calibration process from FP32 model without the need of fine-tuning or retraining. We perform a systematical and comprehensive study on 18 widely-used convolutional neural networks and demonstrate the effectiveness of 8-bit low precision inference across a wide range of applications and use cases, including image classification, object detection, image segmentation, and super resolution. We show that the inference throughput and latency are improved by 1.6X and 1.5X respectively with minimal within 0.6%1to no loss in accuracy from FP32 baseline. We believe the methodology can provide the guidance and reference design of 8-bit low precision inference for other frameworks. All the code and models will be publicly available soon.
Tasks	Calibration, Image Classification, Object Detection, Semantic Segmentation, Super-Resolution
Published	2019-05-01
URL	https://openreview.net/forum?id=SklzIjActX
PDF	https://openreview.net/pdf?id=SklzIjActX
PWC	https://paperswithcode.com/paper/highly-efficient-8-bit-low-precision
Repo
Framework

Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics


Title	Proceedings of the Workshop on Cognitive Modeling and Computational Linguistics
Authors
Abstract
Tasks
Published	2019-06-01
URL	https://www.aclweb.org/anthology/W19-2900/
PDF	https://www.aclweb.org/anthology/W19-2900
PWC	https://paperswithcode.com/paper/proceedings-of-the-workshop-on-cognitive
Repo
Framework

Cross-Domain Training for Goal-Oriented Conversational Agents


Title	Cross-Domain Training for Goal-Oriented Conversational Agents
Authors	Alex Bod{^\i}rl{\u{a}}u, ra Maria, Stefania Budulan, Traian Rebedea
Abstract	Goal-Oriented Chatbots in fields such as customer support, providing certain information or general help with bookings or reservations, suffer from low performance partly due to the difficulty of obtaining large domain-specific annotated datasets. Given that the problem is closely related to the domain of the conversational agent and the data belonging to a specific domain is difficult to annotate, there have been some attempts at surpassing these challenges such as unsupervised pre-training or transfer learning between different domains. A more thorough analysis of the transfer learning mechanism is justified by the significant improvement of the results demonstrated in the results section. We describe extensive experiments using transfer learning and warm-starting techniques with improvements of more than 5{%} in relative percentage of success rate in the majority of cases, and up to 10x faster convergence as opposed to training the system without them.
Tasks	Transfer Learning
Published	2019-09-01
URL	https://www.aclweb.org/anthology/R19-1017/
PDF	https://www.aclweb.org/anthology/R19-1017
PWC	https://paperswithcode.com/paper/cross-domain-training-for-goal-oriented
Repo
Framework

Connecting the Dots: Learning Representations for Active Monocular Depth Estimation


Title	Connecting the Dots: Learning Representations for Active Monocular Depth Estimation
Authors	Gernot Riegler, Yiyi Liao, Simon Donne, Vladlen Koltun, Andreas Geiger
Abstract	We propose a technique for depth estimation with a monocular structured-light camera, i.e., a calibrated stereo set-up with one camera and one laser projector. Instead of formulating the depth estimation via a correspondence search problem, we show that a simple convolutional architecture is sufficient for high-quality disparity estimates in this setting. As accurate ground-truth is hard to obtain, we train our model in a self-supervised fashion with a combination of photometric and geometric losses. Further, we demonstrate that the projected pattern of the structured light sensor can be reliably separated from the ambient information. This can then be used to improve depth boundaries in a weakly supervised fashion by modeling the joint statistics of image and depth edges. The model trained in this fashion compares favorably to the state-of-the-art on challenging synthetic and real-world datasets. In addition, we contribute a novel simulator, which allows to benchmark active depth prediction algorithms in controlled conditions.
Tasks	Depth Estimation, Monocular Depth Estimation
Published	2019-06-01
URL	http://openaccess.thecvf.com/content_CVPR_2019/html/Riegler_Connecting_the_Dots_Learning_Representations_for_Active_Monocular_Depth_Estimation_CVPR_2019_paper.html
PDF	http://openaccess.thecvf.com/content_CVPR_2019/papers/Riegler_Connecting_the_Dots_Learning_Representations_for_Active_Monocular_Depth_Estimation_CVPR_2019_paper.pdf
PWC	https://paperswithcode.com/paper/connecting-the-dots-learning-representations
Repo
Framework

O2U-Net: A Simple Noisy Label Detection Approach for Deep Neural Networks


Title	O2U-Net: A Simple Noisy Label Detection Approach for Deep Neural Networks
Authors	Jinchi Huang, Lie Qu, Rongfei Jia, Binqiang Zhao
Abstract	This paper proposes a novel noisy label detection approach, named O2U-net, for deep neural networks without human annotations. Different from prior work which requires specifically designed noise-robust loss functions or networks, O2U-net is easy to implement but effective. It only requires adjusting the hyper-parameters of the deep network to make its status transfer from overfitting to underfitting (O2U) cyclically. The losses of each sample are recorded during iterations. The higher the normalized average loss of a sample, the higher the probability of being noisy labels. O2U-net is naturally compatible with active learning and other human annotation approaches. This introduces extra flexibility for learning with noisy labels. We conduct sufficient experiments on multiple datasets in various settings. The experimental results prove the state-of-the-art of O2S-net.
Tasks	Active Learning
Published	2019-10-01
URL	http://openaccess.thecvf.com/content_ICCV_2019/html/Huang_O2U-Net_A_Simple_Noisy_Label_Detection_Approach_for_Deep_Neural_ICCV_2019_paper.html
PDF	http://openaccess.thecvf.com/content_ICCV_2019/papers/Huang_O2U-Net_A_Simple_Noisy_Label_Detection_Approach_for_Deep_Neural_ICCV_2019_paper.pdf
PWC	https://paperswithcode.com/paper/o2u-net-a-simple-noisy-label-detection
Repo
Framework