Table 3 Comparison of reconstructions on the event camera dataset in terms of PSNR and SSIM.
From: An event-oriented diffusion-refinement method for sparse events completion
Scenes | PSNR\(\uparrow\) | SSIM\(\uparrow\) | ||||||
|---|---|---|---|---|---|---|---|---|
STCL | PoinTr | VRCNet | Ours | STCL | PoinTr | VRCNet | Ours | |
boxes_6dof | 20.50 | 8.91 | 21.39 | 21.69 | 0.33 | 0.46 | 0.36 | 0.46 |
calibration | 17.70 | 14.97 | 17.37 | 18.24 | 0.41 | 0.50 | 0.46 | 0.58 |
dynamic_6dof | 16.68 | 16.17 | 17.66 | 18.87 | 0.47 | 0.67 | 0.45 | 0.70 |
office_zigzag | 20.68 | 15.67 | 16.29 | 17.56 | 0.56 | 0.59 | 0.45 | 0.61 |
poster_6dof | 19.47 | 8.81 | 19.36 | 19.87 | 0.37 | 0.52 | 0.39 | 0.54 |
shapes_6dof | 21.11 | 18.57 | 19.58 | 17.42 | 0.80 | 0.82 | 0.83 | 0.88 |
slider_depth | 22.46 | 15.30 | 16.15 | 21.11 | 0.69 | 0.67 | 0.55 | 0.74 |
Average | 19.42 | 10.96 | 19.65 | 20.23 | 0.40 | 0.54 | 0.41 | 0.55 |
