The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.
Thank you for the great work! While attempting to reproduce the evaluation results, I noticed some inconsistency in the video samples (Table 16) between the datasets LVOSv2 and VOST and the ones used in the paper. Specifically, the LVOSv2 validation set originally contains 140 videos, but only 136 are sampled in the evaluation. Similarly, the VOST dataset originally had 70 videos, but the evaluation uses 636. Due to these inconsistencies, I am unable to reproduce the evaluation results. Could you provide clarification on this difference? Thanks!
Thank you for the great work! While attempting to reproduce the evaluation results, I noticed some inconsistency in the video samples (Table 16) between the datasets LVOSv2 and VOST and the ones used in the paper. Specifically, the LVOSv2 validation set originally contains 140 videos, but only 136 are sampled in the evaluation. Similarly, the VOST dataset originally had 70 videos, but the evaluation uses 636. Due to these inconsistencies, I am unable to reproduce the evaluation results. Could you provide clarification on this difference? Thanks!