• Publications
  • Influence
Aggregated Residual Transformations for Deep Neural Networks
On the ImageNet-1K dataset, it is empirically show that even under the restricted condition of maintaining complexity, increasing cardinality is able to improve classification accuracy and is more effective than going deeper or wider when the authors increase the capacity. Expand
Holistically-Nested Edge Detection
HED performs image-to-image prediction by means of a deep learning model that leverages fully convolutional neural networks and deeply-supervised nets, and automatically learns rich hierarchical representations that are important in order to resolve the challenging ambiguity in edge and object boundary detection. Expand
Integral Channel Features
It is demonstrated that when designed properly, integral channel features not only outperform other features including histogram of oriented gradient (HOG), they also result in fast detectors when coupled with cascade classifiers. Expand
Deeply Supervised Salient Object Detection with Short Connections
A new saliency method is proposed by introducing short connections to the skip-layer structures within the HED architecture, which produces state-of-the-art results on 5 widely tested salient object detection benchmarks, with advantages in terms of efficiency, effectiveness, and simplicity over the existing algorithms. Expand
Rethinking Spatiotemporal Feature Learning: Speed-Accuracy Trade-offs in Video Classification
It is shown that it is possible to replace many of the 3D convolutions by low-cost 2D convolution, suggesting that temporal representation learning on high-level “semantic” features is more useful. Expand
Detecting texts of arbitrary orientations in natural images
A system which detects texts of arbitrary orientations in natural images using a two-level classification scheme and two sets of features specially designed for capturing both the intrinsic characteristics of texts to better evaluate its algorithm and compare it with other competing algorithms. Expand
Similarity network fusion for aggregating data types on a genomic scale
Similarity network fusion substantially outperforms single data type analysis and established integrative approaches when identifying cancer subtypes and is effective for predicting survival. Expand
Deeply-Supervised Nets
The proposed deeply-supervised nets (DSN) method simultaneously minimizes classification error while making the learning process of hidden layers direct and transparent, and extends techniques from stochastic gradient methods to analyze the algorithm. Expand
Probabilistic boosting-tree: learning discriminative models for classification, recognition, and clustering
  • Zhuowen Tu
  • Mathematics, Computer Science
  • Tenth IEEE International Conference on Computer…
  • 17 October 2005
The applications of PBT for classification, detection, and object recognition are shown and the framework has interesting connections to a number of existing methods such as the A* algorithm, decision tree algorithms, generative models, and cascade approaches. Expand
Auto-Context and Its Application to High-Level Vision Tasks and 3D Brain Image Segmentation
  • Zhuowen Tu, X. Bai
  • Computer Science, Medicine
  • IEEE Transactions on Pattern Analysis and Machine…
  • 1 October 2010
The scope of the proposed algorithm goes beyond image analysis and it has the potential to be used for a wide variety of problems for structured prediction problems, including high-level vision and medical image segmentation problems. Expand