Abstract: Oriented object detection (OOD) in remote sensing images (RSIs) remains a challenging work due to an arbitrary orientation of instances. Learning rotation-invariant features is critical in ...
We introduce TASTE-Rob: 1) a dataset with 100,856 task-oriented hand-object interaction videos, 2) a three-stage pose-refinement video generation pipeline. With the above contributions, TASTE-Rob is ...
What if a device could see the world the same way humans do, seeing objects, recognizing them, and understanding what they are in real time? Just like our eyes capture visuals and our brain instantly ...
Abstract: Recent neural models for video captioning are typically built using a framework that combines a pre-trained visual encoder with a large language model(LLM) decoder. However, large language ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results