Description

Welcome to the 3rd MSU-ND Computer Vision and Biometrics Workshop. This is a gathering of computer vision and biometrics researchers at Michigan State University and the University of Notre Dame to discuss recent research findings and cultivate a community of collaboration among the faculty and students.

Program Outline

Poster Session I

1. InstructCLIP: Learning Instruct-aware Visual Representation (Zhiyuan Ren)
2. Instruction Tuning for Emotion Recognition in LLMs: Challenges and Early Findings (Morgan Sandler)
3. Procedural Mistake Detection via Action Effect Modeling (Wenliang Guo)
4. Uncertainty-Empowered Survival Risk Prediction with Whole Slide Images (Vihanga Jayawickrama)
5. Sign Election and Coordinate-wise Aggregation of Gradients for Byzantine Tolerant Federated Learning (Joseph Geo Benjamin)
6. OASIS Uncovers: High-Quality T2I Models, Same Old Stereotypes (Sepher Dehdashtian)
7. Breaking The Signal: Adversarial Attacks and Defense Mechanisms in Automatic Speech Recognition System (Daniyal Dar)
8. Extreme Value Theory for Modeling Category Decision Boundaries in Visual Recognition (Jin Huang)
9. EyePreserve: Identity-Preserving Iris Synthesis (Siamul Karim Khan)
10. On the Effectiveness of Aggregated Receptive Fields for Interpretable Face Verification (Kağan Öztürk)
11. Saliency-Guided Training of Fingerprint Presentation Attack Detection Models (Samuel Webster)
12. Salient Information-Preserving Adversarial Training (Timothy Redgrave)
13. Remote Photoplethysmography (rPPG) for Privacy-Aware Biometric Systems (Lu Niu)
14. SecureRAG: End-to-End Secure Retrieval-Augmented Generation (Amina Bassit)

Poster Session II

1. A Single Line of Code to Improve Similarity Scores for Open-Set Biometrics (Yiyang Su)
2. Real-Time Assessment of Visual Expertise for Effective Human-AI Teams (Byron Dowling)
3. Generative Zero-Shot Composed Image Retrieval (Wei Ao)
4. Psych-Occlusion: Using Visual Psychophysics for Aerial Detection of Occluded Persons during Search and Rescue (Arturo Russell Bernal)
5. Rethinking Vision-language Model in Face Forensic: Multi-modal Interpretable Forged Face Detector (Xiao Guo)
6. Comparative Medical Radiography using CNNs, Transformers and LLMs (Redwan Sony)
7. Open Set DeepFake Detection with Category-specific Characteristics (Zhongyi Cai)
8. DiverseFlow: Sample-Efficient Diverse Mode Coverage in Flows (Mashrur Morshed)
9. Has the Virtualization of the Face Changed Facial Perception? (Louisa Conwill)
10. MonoCoP: Chain-of-Prediction for Monocular 3D Object Detection. (Zhihao Zhang)
11. CoInD: Enabling Logical Compositions in Diffusion Models (Sachit Gaudi)
12. Biometrics at Life’s Extremes: Infant and Post-Mortem Iris Recognition (Rasel Ahmed Bhuiyan)
13. Encoding and Synthesizing Human Facial Expressions (Hamzeh Alzweri)
14. Erasing The Invisible: Adversarial Vulnerability of GenAI Watermarking (Noor Hussien)

Poster Session III

1. Improved Ear Verification with Vision Transformers and Overlapping Patches (Deeksha Arun)
2. RePLAy: Remove Projective LiDAR Depthmap Artifacts via Exploiting Epipolar Geometry (Girish Chandar G)
3. Vec2Face: Scaling Face Dataset Generation with Loosely Constrained Vectors (Haiyu Wu)
4. ViT-Split: Unleashing the Power of Vision Foundation Models via Efficient Splitting Heads (Yifan Li)
5. Exploring the Viability of Deep Learning Methods for Liveness Detection in Dynamic Vehicle Scenarios (Ben Sporrer)
6. LLM-augmented Disentangled Diffusion Model for Face Generation (Zhizhong Huang)
7. Privacy-Safe Iris Presentation Attack Detection (Mahsa Mitcheff)
8. SEAL: Semantic Attention Learning for Long Video Representation (Lan Wang)
9. Eye Contact: A Dual-Stage Approach for Automated Contact Lens Detection (Parisa Farmanifard)
10. Story Grammar Semantic Matching for Literary Study (Abby Swenor)
11. Subtoken Image Transformer (SiT) for Generalized Category Discovery (Jie Zhu)
12. CRAFT: Contextual Reactivation of Filters for Face Recognition Training (Aman Bhatta)
13. Show Me: Generating Instructional Videos with Diffusion Models (Yujiang Pu)

Parking Information

The workshop will occur in 205 – 207 McKenna Hall on the Notre Dame Campus. Parking is FREE of charge, but note that lots will be busier than usual due to a football event on campus that day. Please follow the directions and instructions of parking attendants, police, and security staff.

• Option 1: Proceed to campus from the south via Edison Road and Angela Blvd. to the parking lots located south of the Joyce Center and Stadium. These lots are expected to fill up quickly in the morning, however.
• Option 2: North of the stadium, parking is available in the Library and D2 North lots. These should be accessed via Twyckenham Parkway. Proceed west on Bulla Road to the Library Lot. D2 North lots are available from Douglas Road and then proceed south on Wilson Drive to the lot.
• Option 3: In addition, the White Field parking area (located west of Juniper Road between Douglas and Pendle Roads) will be available and can be accessed from Cleveland Road via southbound on Juniper Road. Transpo buses will be on hand to shuttle you from White Field to the Library Circle beginning at 7:00 a.m. and until 4:00 p.m. The entrance to the White Field directly from the Toll Road will NOT be open.