zhangxiaosong18 / FreeAnchor

FreeAnchor: Learning to Match Anchors for Visual Object Detection (NeurIPS 2019)

https://arxiv.org/abs/1909.02466

zhangxiaosong18/FreeAnchor Issues

Difference between code and paper
Updated 3 years ago1
RuntimeError when training
Updated 4 years ago6
AP gap of ResNet-50 based FreeAnchor
Updated 4 years ago
【APs】All APs are -1.0 when I test pretrained_models on 'coco_test-dev'
Closed 4 years ago
formulation error
Updated 4 years ago3
【AP】got unexpected AP results
Closed 4 years ago
positive bag loss decrease very quickly
Closed 4 years ago
loss increases when training
Closed 4 years ago
CPU usage goes too high when running demo/webcam.py
Closed 4 years ago9
memory leak in boxlist_iou?
Updated 4 years ago1
import _c error
Updated 4 years ago3
Any step-by-step results available?
Updated 4 years ago
import _c error
Closed 4 years ago2
How do i test 5,000 images?
Closed 4 years ago1
Does FreeAnchor require artificially designed anchor boxes?Such as anchor_rations=[0.5, 1.0, 2.0]
Updated 4 years ago2
RuntimeError: The size of tensor a (67) must match the size of tensor b (68) at non-singleton dimension 3
Updated 4 years ago2
AttributeError: 'tuple' object has no attribute 'values'
Updated 4 years ago1
RuntimeError: CUDA error: an illegal memory access was encountered
Closed 4 years ago2
distributed.deprecated
Closed 4 years ago2
AttributeError: 'Image' object has no attribute 'shape'
Updated 4 years ago1
about test process
Closed 4 years ago1
Undefined name 'mask_utils' in segmentation_mask.py
Closed 4 years ago1
AttributeError: nms
Updated 4 years ago1
Problem about calculation of loss
Closed 4 years ago2
Question about test?
Closed 4 years ago1
can it train on dataset that only has bounding boxes?
Closed 4 years ago1
Does it support train on my own dataset?
Closed 4 years ago3
训练loss的一些问题
Closed 4 years ago2
萌新一名，请问必须要搭建maskrcnn-benchmark的环境吗？ I'm a beginner ，do I have to install the environment of maskrcnn-benchmark
Closed 4 years ago2
RuntimeError: CUDA error: device-side assert triggered
Closed 5 years ago2
Performance gap!
Closed 5 years ago2
positive_loss is much larger than negative_loss
Updated 5 years ago1
Use Group Normalization
Updated 5 years ago
dimension specified as 0 but tensor has no dimensions
Closed 5 years ago5
how long it takes to train coco data?
Updated 5 years ago2
Compile Failed on Windows 10
Closed 5 years ago2