Last active
June 14, 2020 10:14
-
-
Save yucedagonurcan/ed4888b66520f08788a1180b03d961fc to your computer and use it in GitHub Desktop.
Faster R-CNN Map Mask Results
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Faster R-CNN Results | |
| --------------------- | |
| --------------------- | |
| Pure VOC2007: | |
| ----------------------------------------------------------- | |
| Overlap_Coef:0, Top_300, 0.334 secs/image, car mAP: 0.7618 | |
| Overlap_Coef:0, Top_50, 0.260 secs/image, car mAP: 0.7624 | |
| ------------------------------------------------------------ | |
| nuScenes Dataset: | |
| ------------------------------------------------------------------------------------------ | |
| TOP_300 | |
| ------- | |
| +Overlap_Coef:0, Top_300, 1.329 secs/image, car mAP: 0.2868 (Calculate Map Overlap but don't use it) | |
| +Overlap_Coef:0, Top_300, 0.420 secs/image, car mAP: 0.2875 (Don't Calculate Map Overlap) | |
| +Overlap_Coef:1, Top_300, 1.290 secs/image, car mAP: 0.2859 | |
| +Overlap_Coef:2, Top_300, 1.273 secs/image, car mAP: 0.2869 | |
| +Overlap_Coef:3, Top_300, 1.302 secs/image, car mAP: 0.2870 | |
| TOP_50 | |
| ------- | |
| Overlap_Coef:0, 1.271 secs/image, car mAP: 0.2855 (Calculate Map Overlap but don't use it) | |
| Overlap_Coef:0, 0.395 secs/image, car mAP: 0.2853 (Don't Calculate Map Overlap) | |
| Overlap_Coef:1, 1.191 secs/image, car mAP: 0.2775 | |
| Overlap_Coef:2, 1.193 secs/image, car mAP: 0.2509 | |
| Overlap_Coef:3, 1.190 secs/image, car mAP: 0.2505 | |
| TOP_20 | |
| ------- | |
| Overlap_Coef:0, 0.321 secs/image, car mAP: 0.2492 (Calculate Map Overlap but don't use it) | |
| Overlap_Coef:0, 1.174 secs/image, car mAP: 0.2492 (Don't Calculate Map Overlap) | |
| Overlap_Coef:1, 1.177 secs/image, car mAP: 0.2424 | |
| Overlap_Coef:2, 1.184 secs/image, car mAP: 0.2375 | |
| Overlap_Coef:3, 1.160 secs/image, car mAP: 0.2221 | |
| ------------------------------------------------------------------------------------------ | |
| YOLO Results | |
| ------------------------ | |
| ------------------------ | |
| Original mAP for Car: 66.41 | |
| Map Mask Early Fusion mAP for Car: 65.99 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment