A method for object detection by text promptingRot, Žiga (Avtor) Kristan, Matej (Mentor) Pelhan, Jer (Komentor) object detectioncomputer visiondeep learningThe 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DINO is a prominent zero-shot detector, but its training code is not fully open-source, and its implementation is outdated. In this work, we reimplement Grounding DINO, achieving ~20% speedup, and extend it for text-based object counting, though these modifications do not consistently improve FSCD-147 performance. To enable training from scratch, we optimize the model further, achieving an additional ~30% speedup, and develop an edge-oriented variant inspired by a closed-source model. Training on 1.3 million images from scratch, we evaluate the model on COCO and LVIS datasets, where it performs comparably to other open-source models, but remains below the closed-source baseline, likely due to a much smaller training set.20252025-10-14 13:55:01Magistrsko delo/naloga175070sl