1st JHU Computer Vision Workshop by Vision Professors - Drs. Alan Yuille, Vishal Patel, Rene Vidal, Gregory Hager

Event Timing: Friday April 5, 2019 at 8:30 a.m. - 5:00 p.m.

Event Address: Salon C at Charles Commons. It is located at 10 E. 33rd Street, Baltimore 21218. Enter the building through the 33rd street entrance (next to Insomnia Cookies)

Contact us at (410) 516-6745 or yylu1989@gmail.com ; ylu80@jhu.edu ; lmutesi1@jhu.edu ; vishwanath.sindagi@gmail.com

Talks

Research overview by Drs. Alan Yuille, Vishal Patel, Rene Vidal, Gregory Hager.

Oral Presentations

Please keep presentations to 13 minutes talk + 2 min questions in length.

Schedule

8:30-9:00 Breakfast and Opening Remarks. Breakfast will be served.

Research Overview

Time	Speaker
9:00-9:20	Prof. Yuille
9:20-9:40	Prof. Patel
9:40-10:00	Prof. Hager
10:00-10:20	Prof. Vidal

Poster Session I and Coffee Break

10:20-11:20

Oral Session I

Time	Title	Presenter
11:30-11:45	Feature Denoising for Improving Adversarial Robustness	Cihang Xie
11:45-12:00	Deep Transfer Learning for Multiple Class Novelty Detection	Pramuditha Perera
12:00-12:15	SAFER: Fine-grained Activity Detection by Compositional Hypothesis Testing	Tae Soo Kim
12:15-12:30	End-to-End Fine-Grained Action Segmentation and Recognition Using CRF Models and Discriminative Sparse Coding	Effrosyni Mavroudi

12:30-1:30 Lunch break. Lunch will be served.

Oral Session II

Time	Title	Presenter
2:00-2:15	Semi-Supervised 3D Multi-Organ Segmentation	Yuyin Zhou
2:15-2:30	C2AE : Class Conditioned Auto-Encoders for Open-Set Recognition	Poojan Oza
2:30-2:45	Analyzing and Undoing the Damage of Dataset Bias to Face Recognition with Synthetic Faces	Adam Kortylewski
2:45-3:00	Multi-Cell Classification by Convolutional Dictionary Learning with Class Proportion Priors	Florence Yellin
3:00-3:15	Towards Accurate Task Accomplishment with Low-Cost Robotic Arms	Weichao Qiu
3:15-3:30	Representing and Estimating Assembly Processes	Jonathan Jones
3:30-3:45	Recurrent neural networks for classifying human embryonic stem cell-derived cardiomyocytes	Carolina Pacheco

Poster Session II and Coffee Break

3:45-4:45

4:45-5:00 Closing Remarks

10:20-11:20 Poster Session I

Poster ID	Title	Presenter
A1	Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation	Chenxi Liu
A2	Uncertainty guided multi-scale residual learning using a cycle spinning CNN for single image de-raining	Rajeev Yasarla
A3	ELASTIC: Improving CNNs with Instance Specific Scaling Policies	Huiyu Wang
A4	Dual Principal Component Pursuit: Improved Analysis and Efficient Algorithms	Zhihui Zhu
A5	Visual Concepts for CNNs: Towards Unsupervised, Few-shots, Occlusions and Adversaries	Qing Liu
A6	Neural Rejuvenation: Improving Deep Network Training by Enhancing Computational Resource Utilization	Siyuan Qiao
A7	Unsupervised Learning for Surgical Motion by Learning to Predict the Future	Robert DiPietro
A8	An Alarm System For Segmentation Algorithm Based On Shape Model	Fengze Liu
A9	Conformal Symplectic and Relativistic Optimization	Guilherme Franca
A10	Snapshot Distillation: Teacher-Student Optimization in One Generation	Chenglin Yang
A11	DAFE-FD: Density Aware Feature Enrichment for Face Detection	Vishwanath Sindagi
A12	fd-GAN: Pose-guided Feature Distilling GAN for Robust Person Re-identification	Zhuowan Li
A13	Robust Face Detection via Learning Small Faces on Hard Images	Zhishuai Zhang
A14	End-to-End Fine-Grained Action Segmentation and Recognition Using CRF Models and Discriminative Sparse Coding	Effrosyni Mavroudi
A15	Bridging the Gap Between 2D and 3D Organ Segmentation with Volumetric Fusion Net	Yingda Xia
A16	Deep Transfer Learning for Multiple Class Novelty Detection	Pramuditha Perera
A17	SAFER: Fine-grained Activity Detection by Compositional Hypothesis Testing	Tae Soo Kim
A18	Adversarial Examples for Edge Detection: They Exist, and They Transfer	Christian Cosgrove
A19	Feature Denoising for Improving Adversarial Robustness	Cihang Xie
A20	Regional Homogeneity: Towards Learning Transferable Universal Adversarial Perturbations Against Defenses	Yingwei Li
A21	Dropout as a Low-Rank Regularizer for Matrix Factorization	Connor Lane
A22	Septoplasty Skill Validation on Multi-Site Dataset	Matthew Holden

3:45-4:45 Poster Session II

Poster ID	Title	Presenter
B1	Grad-CAM++: Generalized Gradient-based Visual Explanations for Deep Convolutional Networks	Aditya Chattopadhyay
B2	Deep Regression Forests for Age Estimation	Wei Shen
B3	High-Quality Facial Photo-Sketch Synthesis Using Multi-Adversarial Networks	Lidan Wang
B4	Image Generation from Sketch Constraint Using Contextual GAN	Yongyi Lu
B5	Unsupervised Detection of Tool Presence in Endoscopic Video Frames	David Z. Li
B6	Towards Accurate Task Accomplishment with Low-Cost Robotic Arms	Weichao Qiu
B7	Simultaneous Segmentation and Classification of Bone Surfaces from Ultrasound Using a Multi-feature Guided CNN	Puyang Wang
B8	Semi-Supervised 3D Multi-Organ Segmentation	Yuyin Zhou
B9	Thickened 2D Networks for 3D Medical Image Segmentation	Qihang Yu
B10	Deep Hierarchical Multi-label Classification of Chest X-ray Images	Haomin Chen
B11	Surgical Automation from Simulation to Benchtop	Molly O’Brien
B12	Analyzing and Undoing the Damage of Dataset Bias to Face Recognition with Synthetic Faces	Adam Kortylewski
B13	Polarimetric Thermal to Visible Face Verification via Self-Attention Guided Synthesis	Xing Di
B14	SampleAhead: Online Classifier-Sampler Communication for Learning from Synthesized Data	Qi Chen
B15	Recurrent neural networks for classifying human embryonic stem cell-derived cardiomyocytes	Carolina Pacheco
B16	Multi-Cell Classification by Convolutional Dictionary Learning with Class Proportion Priors	Florence Yellin
B17	C2AE : Class Conditioned Auto-Encoders for Open-Set Recognition	Poojan Oza
B18	Representing and Estimating Assembly Processes	Jonathan Jones
B19	OriNet: A Fully Convolutional Network for 3D Human Pose Estimation	Chenxu Luo
B20	Making Deep Neural Network Fooling Practical	Ambar Pal
B21	Computer Vision Techniques for Intraoperative Technical Skill Assessment	Digvijay Singh