Graduate Student Presentations
For CS498GC 4-credit students: You'll give a 15-minute presentation + 5-minute Q&A on an assigned, state-of-the-art SLAM paper.
Please check your assigned SLAM paper and date in the table below.
Please check your assigned SLAM paper and date in the table below.
TL;DR - Key Requirements
- Slides due exactly 1 week before your presentation date (see deadlines below)
- Mandatory TA office hours for slides feedback: attend one week prior to your talk
- Talk length: 15 minutes + 5 minutes Q&A (strict timing; 4 talks/day)
Deadlines at a Glance
| Presentation Date | Slides Due (1 week prior) |
|---|---|
| Fri, Nov 21 | Thu, Nov 14 |
| Wed, Dec 3 | Tue, Nov 26 |
| Fri, Dec 5 | Thu, Nov 28 |
| Wed, Dec 10 | Tue, Dec 3 |
What We Want You to Learn (and Show)
- Methodology & architecture — explain the system diagram and motivate each design choice
- Novelty — what's truly new and why it matters
- Limitations — be specific and evidence-based (failure cases, assumptions, compute/data constraints)
- Improvements — concrete ideas to address those limitations
Prof. Girish's suggested questions for a compelling talk:
- Why is it important?
- Why is it hard?
- How is it novel?
- Why does it work?
Suggested Slide Outline (Target: 15 minutes)
- Problem & motivation (tie explicitly to mobile robotics/SLAM)
- Prior art (2–3 key related works; cite fairly but highlight shortcomings)
- Key contribution & approach (detailed system/architecture diagram with time estimates)
- Technical depth (walk us through the math/algorithms; highlight novel formulations)
- Results (datasets, metrics, comparison tables, visualizations; be critical)
- Demo/video (if available—real-world robot footage is gold!)
- Limitations (specific failure cases from the paper or your analysis)
- Future work & your ideas (thoughtful proposals for addressing those limitations)
- Q&A prep (anticipate technical questions on losses, hyperparameters, training)
Target: ~15 slides for 15 minutes
Grading Rubric
- Technical Understanding (40%)
- Accurate explanation of algorithms
- Depth of insight into methodology
- Critical analysis (not just summary)
- Clarity & Organization (30%)
- Logical flow and narrative
- Clear, uncluttered slides
- Effective diagrams & figures
- Critical Analysis (20%)
- Thoughtful limitations discussion
- Creative future work ideas
- Connecting to broader SLAM/robotics context
- Q&A Performance (10%)
- Demonstrating deep understanding
- Handling questions thoughtfully
- Admitting unknowns gracefully
Paper Assignments
| # | Student Name | Paper Title | Paper Link | Submission Deadline | Presentation Date |
|---|---|---|---|---|---|
| 1 | Aryan Chauhan | MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors | CVPR 2025 | Nov 14 | Nov 21 (Fri) |
| 2 | Karteek Gandiboyina | cuVSLAM: CUDA Accelerated Visual Odometry and Mapping | NVIDIA arXiv Preprint | Nov 14 | Nov 21 (Fri) |
| 3 | Het Patel | Open-Vocabulary Online Semantic Mapping for SLAM | arXiv Preprint 2024 | Nov 26 | Dec 3 (Wed) |
| 4 | Ruben Hernandez | VSS-SLAM: Voxelized Surfel Splatting for Geometrically Accurate SLAM | ICRA 2025 | Nov 26 | Dec 3 (Wed) |
| 5 | Yangkun Liu | SCE-LIO: An Enhanced Lidar Inertial Odometry by Constructing Submap Constraints | IEEE RAL | Nov 26 | Dec 3 (Wed) |
| 6 | Nadeem Mohammed | FMCW-LIO: A Doppler LiDAR-Inertial Odometry | IEEE RAL | Nov 26 | Dec 3 (Wed) |
| 7 | Keisuke Ogawa | WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environments | CVPR 2025 | Nov 28 | Dec 5 (Fri) |
| 8 | Tianyu Ren | DRAWER: Interactive 3D Scene Generation from a Single Image | CVPR 2025 | Nov 28 | Dec 5 (Fri) |
| 9 | William Schafer | GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting | CVPR 2024 | Nov 28 | Dec 5 (Fri) |
| 10 | Yingxue Wang | RTG-SLAM: Real-time 3D Reconstruction at Scale using Gaussian Splatting | ACM SIGGRAPH 2024 | Nov 28 | Dec 5 (Fri) |
| 11 | Tongmiao Xu | Wheat3DGS: In-field 3D Reconstruction, Instance Segmentation and Phenotyping of Wheat Heads with Gaussian Splatting | arXiv Preprint 2024 | Dec 3 | Dec 10 (Wed) |
| 12 | Hariprasad Yuvaraj | Adaptive Mobile Manipulation for Articulated Objects In the Open World | arXiv Preprint 2024 | Dec 3 | Dec 10 (Wed) |
| 13 | Xiayu Zhao | VLFM: Vision-Language Frontier Maps for Zero-Shot Semantic Navigation | ICRA 2024 | Dec 3 | Dec 10 (Wed) |
| 14 | Krishna Teja Kolla | Multimodal Spatial Language Maps for Robot Navigation and Manipulation | IJRR 2025 | Dec 3 | Dec 10 (Wed) |
Paper Categories Summary
All graduate students have been assigned state-of-the-art papers from CVPR 2025, SIGGRAPH 2024, ICRA 2024/2025, and IJRR 2025, covering:
- Visual SLAM: VSS-SLAM, DynaVINS++
- Neural SLAM & 3D Gaussian Splatting: WildGS-SLAM, GS-SLAM, RTG-SLAM
- LiDAR SLAM: SCE-LIO, FMCW-LIO
- Semantic SLAM: Open-Vocabulary Online Semantic Mapping
- Vision-Language Navigation: VLFM
- Multimodal Language Navigation: Spatial Language Maps
- Robotic Applications: DRAWER, Adaptive Mobile Manipulation
- Specialized Applications: Wheat3DGS, cuVSLAM
- Real-Time Systems: MASt3R-SLAM
Presentation Schedule Summary
November 21 (Friday) - Before Thanksgiving
2 Presentations:
- Aryan Chauhan - MASt3R-SLAM
- Karteek Gandiboyina - cuVSLAM
December 3 (Wednesday)
4 Presentations:
- Het Patel - Open-Vocabulary Online Semantic Mapping
- Ruben Hernandez - VSS-SLAM
- Yangkun Liu - SCE-LIO
- Nadeem Mohammed - FMCW-LIO
December 5 (Friday)
4 Presentations:
- Keisuke Ogawa - WildGS-SLAM
- Tianyu Ren - DRAWER
- William Schafer - GS-SLAM
- Yingxue Wang - RTG-SLAM
December 10 (Wednesday) - Last Day of Instruction
4 Presentations:
- Tongmiao Xu - Wheat3DGS
- Hariprasad Yuvaraj - Adaptive Mobile Manipulation
- Xiayu Zhao - VLFM
- Krishna Teja Kolla - Multimodal Spatial Language Maps
Quick Links & Resources
- Presentations Table: papers, dates, deadlines (above)
- Syllabus & weekly schedule (presentation weeks noted)
- Policies (integrity, AI-assistant use, late policy)
- Logistics (meeting times, platforms, grade distribution)
- TA Office Hours: Wed 1:30-2:30 PM @ SC 4407
- Questions: Use Campuswire or email ksa5@illinois.edu
If you have conflicts or accessibility needs, please reach out early via Campuswire or email.
All papers and presentation dates assigned! Start reading your assigned paper early with focus on methodology, loss functions, and novel contributions. Remember to attend TA office hours 1 week before your presentation for slide feedback.