JQ

Hi, sorry, the current version is not our final implementation. Since we construct many ablation studies, the code may be confused. In fact, compared with RAB, using the 5 stacked...

I think this is normal, as tracking FPS on low-spec devices is really low.

Hi, Thanks for your attention. The temporal information we emphasize exists between the entire sequence. The mechanism of SNN allows the temporal information of the entire tracking to be captured....

_im_x_pos_ and _im_x_neg_ are aggregated by recording the spatial positions of all occurred negative and positive events during the period, respectively. The _snn_state _first_ definition in the tracking initialization is...

Thank you for pointing it out. We evaluate all trackers by following [pytracking](https://github.com/visionml/pytracking). We checked the checkpoint.pth we uploaded, and the performance on FE240hz is: RSR 60.2%, RPR 87.9%. In...

Hi, We use the same data augmentations as [SiamFC++](https://arxiv.org/abs/1911.06188)(i.e., random shifting and scaling following a uniform distribution on the search image). All compared trackers used their default settings.

Hi, We evaluate all trackers against [pytracking](https://github.com/visionml/pytracking). The _Analysis_ section in the [Readme](https://github.com/visionml/pytracking) provides how to evaluate the tracker.

The link can be opened successfully. Have a try.

Looks like the path is wrong, you should check your local image path.

> Hi guys, please check this repo from Timo&Cedric. It has a trainable e2vid model, but I'm not sure if it is exact the same as e2vid. https://github.com/TimoStoff/event_cnn_minimal @LarryDong hi,...