… the use of object spatial relationship modeling for image captioning, specifically within the
… For image captioning, our architecture uses the feature vectors from the object detector as …

… of objects in an image. Due to the aforementioned reason, we combine object detection with
image captioning to … Initially, we use an object detection model to detect objects in the image …

… In this paper, we validate our proposed method on the VL task of image captioning. By …
-trained object detector described above, our method improves one of the SoTA image captioning …

… object region, not all the motion features. Therefore, object detection that detects an object
region in an image … our proposed method improves image captioning performance. Previous …

… purely as a language modeling problem and presume the availability of a pre-trained fully-supervised
object detector over all object classes of interest, to the best of our knowledge. …

… In order to take advantage of the image classification features and the object detection
features, we add this concatenation step, where we attach the output of the YOLOv4 subsystem as …

… propose CapSal, a salient object detection framework that exploits image captioning to pro-
… Inspired by the success of these works, we propose to leverage image captioning as an …

… Dataset Analysis In this section, we compare our nocaps benchmark to COCO Captions [5]
in terms of both image content and caption diversity. Based on ground-truth object detection …

… Two stage attention models consists of bottom-up attention and top-down attention, where
bottom-up attention first uses object detection models to detect multiple informative regions in …

… object detectors using bounding box annotations for a limited set of object categories, as well
as image… problem of knowledge transfer from image-caption pretraining to object detection. …