DeepView'21 @ AVSS

Call for Papers

We are soliciting high quality papers covering the topics listed below. Papers should follow the standard AVSS formatting instructions. Paper length should be 4 to 8 pages according to the AVSS format. Accepted papers will appear in the AVSS proceedings.

• Submission Deadline: September 01, 2021 → October 03, 2021 → October 13, 2021 → October 20, 2021 → October 27, 2021 (No more extension of deadline)

• Author Notification: September 30, 2021 → October 08, 2021 → October 16, 2021 → October 23, 2021 → October 30, 2021 → November 1, 2021 (No more extension of deadline)

• Camera Ready Due: October 08, 2021 → October 15, 2021 → October 22, 2021 → October 29, 2021 → November 5, 2021 (No more extension of deadline)

• Submission via CMT: https://cmt3.research.microsoft.com/DeepView2021

• Workshop day: November 16, 2021

Topics of Interest

We embrace the most advanced deep learning systems based on real-time large-scale analysis, meanwhile being open to classical physically grounded models and feature engineering, as well as any well-motivated combination of the two streams. We will solicit papers from but not limited to the following topics:

• Robust recognition and detection in the surveillance videos

• Fast Large-scale multi-camera multi-object application (detection, tracking, re-identification, and others)

• 2D/3D pose estimation in the surveillance videos

• One/few shot learning in the unconstrained surveillance scenarios

• Anomaly detection in the surveillance videos

• Human behavior analysis and recognition in the surveillance videos

• Generation of visual data for surveillance analysis system

• Privacy-preserving visual learning

• Applications and systems for security and safe

Accepted Presentations are as follows:

1. On the Performance of Crowd-Specific Detectors in Multi-Pedestrian Tracking

Daniel Stadler* (Karlsruhe Institute of Technology) and Jürgen Beyerer (Fraunhofer IOSB)

2. Multi-Pedestrian Tracking with Clusters;

Daniel Stadler* (Karlsruhe Institute of Technology) and Jürgen Beyerer (Fraunhofer IOSB

3. Unsupervised Vehicle Re-Identification via Self-supervised Metric Learning using Feature Dictionary;

Jongmin Yu* and Hyeontaek Oh (KAIST)

4. Object Detection Considering Anchor Shape And Image Deblurring Method for Real-World Applications;

Sungjin Cho*, Soyeol Lee (Korea University) Jinho Yoo (Sogang University), Seung-Jin Baek and Sung-Jea Ko (Korea University)

5. Task-Driven Deep Image Enhancement Network for Autonomous Driving in Bad Weather

Younkwan Lee*, Jihyo Jeon, Yeongmin Ko, Byung-Gwan Jeon and Moongu Jeon (GIST).