faster-rcnn训练成功

最新推荐文章于 2024-12-13 17:39:51 发布

原创最新推荐文章于 2024-12-13 17:39:51 发布 · 776 阅读

0 ·

CC 4.0 BY-SA版权

Deep Learning 专栏收录该内容

172 篇文章

订阅专栏

本文记录了Faster R-CNN算法在VOC2007数据集上的测试结果，详细展示了各类目标检测的平均精度（AP），包括飞机、自行车、汽车等，并给出了平均精度（mAP）为0.5914。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

Saving cached annotations to /home/faster-rcnn/py-faster-rcnn/data/VOCdevkit2007/annotations_cache/annots.pkl
AP for aeroplane = 0.5994
AP for bicycle = 0.7278
AP for bird = 0.5605
AP for boat = 0.4293
AP for bottle = 0.3362
AP for bus = 0.6674
AP for car = 0.7399
AP for cat = 0.7064
AP for chair = 0.3685
AP for cow = 0.6038
AP for diningtable = 0.5896
AP for dog = 0.6620
AP for horse = 0.7638
AP for motorbike = 0.6906
AP for person = 0.6588
AP for pottedplant = 0.3292
AP for sheep = 0.5628
AP for sofa = 0.5275
AP for train = 0.6898
AP for tvmonitor = 0.6145
Mean AP = 0.5914
~~~~~~~~
Results:
0.599
0.728
0.560
0.429
0.336
0.667
0.740
0.706
0.369
0.604
0.590
0.662
0.764
0.691
0.659
0.329
0.563
0.528
0.690
0.615
0.591
~~~~~~~~

--------------------------------------------------------------
Results computed with the **unofficial** Python eval code.
Results should be very close to the official MATLAB eval code.
Recompute with `./tools/reval.py --matlab ...` for your paper.
-- Thanks, The Management
--------------------------------------------------------------

real 3m20.092s
user 3m44.560s
sys 0m25.752s