ClipCrop: Conditioned Cropping Driven by Vision-Language Model | IEEE Conference Publication | IEEE Xplore