From Local Matches to Global Masks: Template-Guided Instance Detection and Segmentation in Open-World Scenes
cs.CV, cs.RO
Submitted: 2026-03-03
Updated: 2026-05-13
Code: https://github.com/ros-planning/navigation
Project page: https://irvlutd.github.io/L2G
Terminology
Sources
- Build Once, Monitor Continuously: Persistent Semantic Mapping via Autonomous Exploration and Open-Vocabulary Object Updates
- Target Driven Instance Detection
- Perception Encoder: The best visual embeddings are not at the output of the network
- SAM 3: Segment Anything with Concepts
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- Adam: A Method for Stochastic Optimization
- Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection
- CLIP-Adapter: Better Vision-Language Models with Feature Adapters
- Representation Learning with Contrastive Predictive Coding
- DINOv2: Learning Robust Visual Features without Supervision
- SAM 2: Segment Anything in Images and Videos
- DINOv3
- TS-SAM: Fine-Tuning Segment-Anything Model for Downstream Tasks
- Objects as Points
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs