Search papers, labs, and topics across Lattice.
The paper introduces UHR-BAT, a novel vision-language model designed for efficient processing of ultra-high-resolution remote sensing imagery by compressing visual tokens under a strict computational budget. It uses text-guided, multi-scale importance estimation to identify and retain query-critical tokens while discarding redundant ones. Experiments demonstrate state-of-the-art performance on various benchmarks, showing that UHR-BAT effectively balances detail preservation and computational efficiency.
Achieve state-of-the-art performance on ultra-high-resolution remote sensing tasks without the quadratic compute cost, thanks to a query-guided token compression strategy.
Ultra-high-resolution (UHR) remote sensing imagery couples kilometer-scale context with query-critical evidence that may occupy only a few pixels. Such vast spatial scale leads to a quadratic explosion of visual tokens and hinders the extraction of information from small objects. Previous works utilize direct downsampling, dense tiling, or global top-k pruning, which either compromise query-critical image details or incur unpredictable compute. In this paper, we propose UHR-BAT, a query-guided and region-faithful token compression framework to efficiently select visual tokens under a strict context budget. Specifically, we leverage text-guided, multi-scale importance estimation for visual tokens, effectively tackling the challenge of achieving precise yet low-cost feature extraction. Furthermore, by introducing region-wise preserve and merge strategies, we mitigate visual token redundancy, further driving down the computational budget. Experimental results show that UHR-BAT achieves state-of-the-art performance across various benchmarks. Code will be available at https://github.com/Yunkaidang/UHR.
