Get Started
Home
Topics
Search
Library
Research questionHow can visual counting represent both individual objects and semantic groups within the same image?An image may contain objects that should be counted individually as well as meaningful groups such as bunches, stacks, or pairs. A counting system must therefore determine not only what category is relevant but also which semantic unit the count refers to.
AI
Computer Vision
Evaluation & Benchmarks
Image & Video Processing
Machine Learning
Research Paper
Latest papersRecent research connected to this question, newest first.Counting Beyond Instances: A Benchmark for Group-Individual Object CountingThe source defines a group-and-individual counting task, supplies paired annotations and containment relations in the BunchCount benchmark, and evaluates a relational counting framework for real-world still images. Its evidence is limited to that benchmark and task formulation.research paper · Sep 4, 2026
Related questions
How can cooperative semantic communication jointly support classification and regression in visual perception?How can we semantically compare autonomous-driving image subsets at scale and attribute their differences to objects?How should self-supervised visual learning combine objectives to prevent collapse while preserving semantic and spatial information?How can computer-vision systems use commonsense to reason about scenes, object relations, and actions?