Derek Hoiem
Professor
Studies computer vision, including scene understanding, object recognition and general-purpose vision systems.
Derek Hoiem's lab at UIUC works on recognition, detection & segmentation and data management & big data. 21 papers since 2023, 13 at top venues, mostly ECCV and 3DV. 5 PhD students and 1 master's student. 5 alumni, most often to Amazon and Google DeepMind. Applications close .
At a glance
- papers, 2023–2026
- 5/ year
- at T1 venues
- 62%
- PhD students
- 5
- MS students
- 1
- alumni listed
- 5
Papers by yearfrom the lab homepage, 2023–2026
- 4 papers2026 · T1 3 · collecting
- 7 papers2025 · T1 4
- 8 papers2024 · T1 5
- 2 papers2023 · T1 1
T1 venuesOtherVenues: ECCV, 3DV, CVPR, NeurIPS, TMLR
Research focuspaper titles that use each topic's words
- object2 papers
- Recognition, detection & segmentation1 paper
- Data management & big data0 papers
- object recognition0 papers
- scene understanding0 papers
Paper list
From the lab homepage.
20264 papers
- ECCV (T1)How to Teach Large Multimodal Models New Skills
- ECCV (T1)SceneDiff: A Benchmark and Method for Multiview Object Change Detection
- ECCV (T1)T-REN: Learning Text-Aligned Region Tokens Improves Dense Vision-Language Alignment and Scalability
- COLMWhy Fine-Tuning Encourages Hallucinations and How to Fix It
20257 papers
- 3DVMonoPatchNeRF: Improving Neural Radiance Fields with Patch-based Monocular Guidance
- NeurIPS (T1)PARTONOMY: Large Multimodal Models with Part-Level Visual Understanding
- 3DVPlenoptic png: Real-time neural radiance fields in 150 kb
- CVPR (T1)RELOCATE: A simple training-free baseline for visual query localization using region-based representations
- NeurIPS (T1)REN: Fast and Efficient Region Encodings from Patch-Based Image Encoders
- TMLR (T1)TextRegion: Text-Aligned Region Tokens from Frozen Image-Text Models
- Findings of EMNLPVisual Program Distillation with Template-Based Augmentation
20248 papers
- 3DVAddressing Low-Shot MVS by Detecting and Completing Planar Surfaces
- ECCV (T1)Anytime Continual Learning for Open Vocabulary Classification
- WACVConsistent Multimodal Generation via A Unified GAN Framework
- TMLR (T1)Continual Learning in Open-vocabulary Classification with Complementary Memory Systems
- CVPR (T1)Region-Based Representations Revisited
- NeurIPS (T1)Stylegan knows normal, depth, albedo, and more
- CVPR (T1)Unified-IO 2: Scaling Autoregressive Multimodal Models with Vision, Language, Audio, and Action
- Findings of NAACLWebWISE: Unlocking Web Interface Control for LLMs via Sequential Exploration
20232 papers
- Journal of Vision 23 (9)Comparing Human Object Learning with Deep Neural Networks
- EMNLP (T1)ViStruct: Visual Structural Knowledge Extraction via Curriculum Guided Code-Vision Representation
Where alumni went5 listed, 100% to industry
Industry 3Academia 0
Most often to
- Amazon
- Google DeepMind
- Salesforce
Apply to UIUCFall 2027
UIUC Siebel Computer Science PhD
- Round
- Single round
- Opens
- —
- GRE
- optional
- English
- TOEFL iBT / IELTS
- Fee
- USD 90
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USUIUC
Svetlana Lazebnik
- Vision-language & multimodal
- CNPeking
Chao Xu
- Video understanding
- CHEPFL
Mathieu Salzmann
- 3D vision & neural rendering
- USUT Austin
Chen Yu
- Video understanding
- USCMU
Alexander Hauptmann
- Video understanding
- GBImperial
Amir Bar
- Generative & foundation models
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
UIUCcloses