The repository contains official Jittor implementations of training and evaluation codes and pre-trained models for SegNext.
The paper is in Here.
Notes: ImageNet Pre-trained models can be found in Jittor Hub mscan_t,mscan_s, mscan_b, mscan_l.
Method | Backbone | Pretrained | Iters | mIoU(ss/ms) | Params | FLOPs | Config | Download |
---|---|---|---|---|---|---|---|---|
SegNeXt | MSCAN-T | IN-1K | 160K | 41.4/42.2 | 4M | 7G | config | Jittor Hub |
SegNeXt | MSCAN-S | IN-1K | 160K | 44.3/45.8 | 14M | 16G | config | Jittor Hub |
SegNeXt | MSCAN-B | IN-1K | 160K | 48.5/49.9 | 28M | 35G | config | Jittor Hub |
SegNeXt | MSCAN-L | IN-1K | 160K | 51.0/52.1 | 49M | 70G | config | Jittor Hub |
Method | Backbone | Pretrained | Iters | mIoU(ss/ms) | Params | FLOPs | Config | Download |
---|---|---|---|---|---|---|---|---|
SegNeXt | MSCAN-T | IN-1K | 160K | 79.8/81.4 | 4M | 56G | config | Jittor Hub |
SegNeXt | MSCAN-S | IN-1K | 160K | 81.3/82.7 | 14M | 125G | config | Jittor Hub |
SegNeXt | MSCAN-B | IN-1K | 160K | 82.6/83.8 | 28M | 276G | config | Jittor Hub |
SegNeXt | MSCAN-L | IN-1K | 160K | 83.2/83.9 | 49M | 578G | config | Jittor Hub |
Method | Backbone | Pretrained | Iters | mIoU(ss) | Params | FLOPs | Config | Download |
---|---|---|---|---|---|---|---|---|
SegNeXt | MSCAN-T | IN-1K | 160K | 68.3 | 4M | - | config | Jittor Hub |
SegNeXt | MSCAN-S | IN-1K | 160K | 68.8 | 14M | - | config | Jittor Hub |
SegNeXt | MSCAN-B | IN-1K | 160K | 69.9 | 28M | - | config | Jittor Hub |
SegNeXt | MSCAN-L | IN-1K | 160K | 70.3 | 49M | - | config | Jittor Hub |