Open-Vocabulary Object Detection
An object detection paradigm in which the model sees only a subset of categories during training and must detect target categories defined by arbitrary textual descriptions at inference time; the core challenge is mapping text semantics to visual regions.
Timeline (last 90 days)
The core challenge of open-vocabulary object detection (OVOD) is that the model must recognize categories it has never seen during training.
VLM-Seek belongs to open-vocabulary object detection, which is fundamentally different from fixed-category detectors such as YOLO and Faster R-CNN.
All Facts (2)
The core challenge of open-vocabulary object detection (OVOD) is that the model must recognize categories it has never seen during training.
50%UnverifiedVLM-Seek belongs to open-vocabulary object detection, which is fundamentally different from fixed-category detectors such as YOLO and Faster R-CNN.
50%