Program

3rd WACV Workshop on Real-World Surveillance (RWS): Applications and Challenges
January 3rd 2023


09:00-09:05 Opening

09:05-09:35 Invited talk 1: Analytics performance in challenging video surveillance scenes
Mats Thulin, AXIS Communications AB, Sweden.

09:35-09:45, paper_7, Synthehicle: Multi-Vehicle Multi-Camera Tracking in Virtual Cities
09:45-09:55, paper_9, Exploiting Inter-pixel Correlations in Unsupervised Domain
Adaptation for Semantic Segmentation
09:55-10:05, paper_11, Mixture Domain Adaptation to Improve Semantic Segmentation in
Real-World Surveillance
10:05-10:15, paper_14, Multimodal Data Augmentation for Visual-Infrared Person ReID with
Corrupted Data
10:15-10:25, paper_15, Transformer Based Multi-Grained Features for Unsupervised Person
Re-Identification
10:25-10:35, paper_16, EMOD: Efficient Moving Object Detection via Image Eccentricity
Analysis and Sparse Neural Networks

10:35-10:45 Coffee Break


10:45-11:15 Invited talk 2: Video surveillance with trustworthiness built-in, Cees G.M.
Snoek
, Netherlands.

11:15-11:25, paper_18, Low-Resolution Thermal Sensor-Guided Image Synthesis
11:25-11:35, paper_19, Knowledge-based Visual Context-Aware Framework for Applications
in Robotic Services
11:35-11:45, paper_21, Exploiting Temporal Context for Tiny Object Detection
11:45-11:55, paper_26, Multi-view Target Transformation for Pedestrian Detection

11:55-12:25 Invited talk 3: Privacy-preserving video analysis, Chen Chen, UCF, USA.

12:25-14:00 Lunch break


14:00-14:30 Invited talk 4: Deep Intermodal Video Analytics, Jack Cooper, IARPA, USA.

14:10-14:20, paper_29, A Transformer-based Late-Fusion Mechanism for Fine-Grained
Object Recognition in Videos
14:20-14:30, paper_34, Bringing Generalization to Deep Multi-View Pedestrian Detection
14:30-14:40, paper_36, Meet-in-the-middle: Multi-scale upsampling and matching for cross-
resolution face recognition
14:40-14:50, paper_37, ThermalSynth: A Novel Approach for Generating Synthetic Thermal
Human Scenarios14:50-15:00, paper_39, An Automated and Scalable ML Solution for Mapping Invasive
Species: the Case of the Australian Tree Fern in Hawaiian Forests

15:00-15:30 Coffee break


15:30-16:00 Invited talk 5: Legal and Ethical Considerations in Urban Reconnaissance
through Supervised Autonomy (URSA), Bart Russell, DARPA, USA.

16:00-16:10, paper_40, VDiSC: An Open Source Framework for Distributed Smart City Vision
and Biometric Surveillance Networks
16:10-16:20, paper_41, Discriminative Sampling of Proposals in Self-Supervised
Transformers for Weakly Supervised Object Localization
16:20-16:30, paper_44, UPAR Challenge: Pedestrian Attribute Recognition and Attributed-
based Person Retrieval – Dataset, Design, and Results


Best Paper Award

The following paper has been chosen to receive the best paper award, sponsored by Milestone Systems A/S:

Inseop Chung, Jayeon Yoo, Nojun Kwak, Exploiting Inter-pixel Correlations in Unsupervised Domain Adaptation for Semantic Segmentation