Search papers, labs, and topics across Lattice.
RDVSv2 is a comprehensive benchmark for RGB-D video salient object detection, featuring 249 video sequences and 29,077 densely annotated frames with depth maps and salient object masks guided by eye-tracking. This dataset addresses the limitations of existing datasets by providing higher quality annotations and more geometry-consistent depth cues, making it significantly larger and more diverse. The authors establish a strong baseline using a parameter-efficient fine-tuning strategy on the Segment Anything Model 2 (SAM2), achieving state-of-the-art performance on RDVSv2 and existing benchmarks, thus highlighting the dataset's challenging nature for current methods.
RDVSv2 reveals that existing RGB-D VSOD methods struggle significantly with its challenging scenarios, setting a new standard in the field.
We introduce RDVSv2, a large-scale benchmark for RGB-D video salient object detection (RGB-D VSOD) with dense frame-level annotations. Existing datasets in this emerging field are often limited in scale and annotation quality, while also relying on less geometry-consistent depth cues. To address these limitations, RDVSv2 is built from publicly accessible stereoscopic online videos and contains 249 video sequences with 29,077 annotated frames. It includes depth maps derived from stereoscopic videos, together with frame-wise salient object masks annotated with eye-tracking guidance. Compared with existing datasets, RDVSv2 is much larger in scale and covers more diverse and challenging scenarios. In addition, we establish a strong baseline for RGB-D VSOD based on Segment Anything Model 2 (SAM2). Specifically, we employ a parameter-efficient fine-tuning (PEFT) strategy to adapt the SAM2 encoder to jointly encode RGB, depth, and optical flow cues. Extensive experiments show that RDVSv2 is substantially more challenging for existing RGB-D VSOD methods. Meanwhile, the proposed baseline achieves state-of-the-art results on RDVSv2 and existing RGB-D VSOD benchmarks. We hope that RDVSv2 and the provided baseline will serve as useful resources for future research on RGB-D VSOD and related multi-modal video understanding tasks. Our dataset and code will be available at https://github.com/ltynick/RDVSv2.
