Nacionalni portal odprte znanosti

Visual object tracking has recently shifted towards target segmentation, which has increased the demand for video datasets with objects segmented in each frame. However, manually obtaining large segmented video datasets is time-consuming and costly. We address this problem by introducing a Semi-supe ...

We tackle the problem of few-shot object counting and detection of arbitrary object categories using only a small number of annotated instances, namely exemplars. The task of the method is to count and detect all objects that are part of the same semantic category as the exemplars but may vary widel ...

V tej nalogi predstavljamo sistem za generiranje učnih podatkov, namenjen izboljšanju učenja modelov za štetje objektov. Obstoječe zbirke pogosto vsebujejo nepravilne ali manjkajoče anotacije ter so večinoma omejene na prevladujoče kategorije objektov. Posledično se modeli pogosto naučijo preštevati ...

The increasing diversity and scale of object detection datasets have highlighted the limitations of closed-set detectors with fixed vocabularies. Open-set object detection addresses this by enabling the detection of arbitrary classes via text prompts. Grounding DINO is a prominent zero-shot detector ...