• Publications
  • Influence
Understanding Convolution for Semantic Segmentation
TLDR
We propose a method called dense upsampling convolution (DUC) to improve pixel-wise semantic segmentation by manipulating convolution-related operations that are of both theoretical and practical value. Expand
  • 553
  • 57
  • PDF
HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-Scale Point Clouds
TLDR
We present a novel deep neural network architecture for end-to-end scene flow estimation that directly operates on large-scale 3D point clouds. Expand
  • 38
  • 12
  • PDF
Grid-GCN for Fast and Scalable Point Cloud Learning
TLDR
This paper introduces Grid-GCN, an approach that blends the advantages of volumetric models and point-based models for fast and scalable point cloud learning. Expand
  • 17
  • 2
  • PDF
Central and peripheral vision for scene recognition: A neurocomputational modeling exploration.
TLDR
We show that the advantage of peripheral vision in scene recognition, as well as the efficiency advantage for central vision, can be replicated using state-of-the-art deep neural network models. Expand
  • 20
  • 1
  • PDF
Are Face and Object Recognition Independent? A Neurocomputational Modeling Exploration
Are face and object recognition abilities independent? Although it is commonly believed that they are, Gauthier et al. [Gauthier, I., McGugin, R. W., Richler, J. J., Herzmann, G., Speegle, M., &Expand
  • 15
  • PDF
Encoding Voxels with Deep Learning
TLDR
Humans achieve fast and accurate recognition of complex objects through the ventral visual stream, a system of interconnected brain regions capable of hierarchical processing of increasingly complex features. Expand
  • 6
  • PDF
A Computational Model of the Development of Hemispheric Asymmetry of Face Processing
TLDR
A Computational Model of the Development of Hemispheric Asymmetry of Face Processing Panqu Wang (pawang@ucsd.edu) Department of Electrical and Computer Engineering, University of California San Diego 9500 Gilman Dr 0407, La Jolla, CA 92093 USA Garrison Cottrell (gary@ucSD.edu). Expand
  • 12
  • PDF
Basic Level Categorization Facilitates Visual Object Recognition
TLDR
We propose a network optimization strategy inspired by both of the developmental trajectory of children's visual object recognition capabilities, and Bar (2003), who hypothesized that basic level information is carried in the fast magnocellular pathway through the prefrontal cortex (PFC) and then projected back to inferior temporal cortex (IT), where subordinate level categorization is achieved. Expand
  • 10
  • PDF
Modeling the Object Recognition Pathway: A Deep Hierarchical Model Using Gnostic Fields
TLDR
In this paper, we use a hierarchical Independent Components Analysis (ICA) algorithm to automatically learn the visual features that account for early visual cortex. Expand
  • 5
  • PDF
Experience Matters: Modeling the Relationship Between Face and Object Recognition
Experience Matters: Modeling the Relationship Between Face and Object Recognition Panqu Wang (pawang@ucsd.edu) Department of Electrical and Computer Engineering, University of California San DiegoExpand
  • 5
  • PDF