Fix bug of FCOS when image has no positive sample (#1136)

* Fix bug of FCOS when image has no positive sample

* somes changes for fcos

mmdet/models/anchor_heads/fcos_head.py

@@ -160,11 +160,11 @@ class FCOSHead(nn.Module):
             avg_factor=num_pos + num_imgs)  # avoid num_pos is 0
 
         pos_bbox_preds = flatten_bbox_preds[pos_inds]
-        pos_bbox_targets = flatten_bbox_targets[pos_inds]
         pos_centerness = flatten_centerness[pos_inds]
-        pos_centerness_targets = self.centerness_target(pos_bbox_targets)
 
         if num_pos > 0:
+            pos_bbox_targets = flatten_bbox_targets[pos_inds]
+            pos_centerness_targets = self.centerness_target(pos_bbox_targets)
             pos_points = flatten_points[pos_inds]
             pos_decoded_bbox_preds = distance2bbox(pos_points, pos_bbox_preds)
             pos_decoded_target_preds = distance2bbox(pos_points,
@@ -339,6 +339,9 @@ class FCOSHead(nn.Module):
     def fcos_target_single(self, gt_bboxes, gt_labels, points, regress_ranges):
         num_points = points.size(0)
         num_gts = gt_labels.size(0)
+        if num_gts == 0:
+            return gt_labels.new_zeros(num_points), \
+                   gt_bboxes.new_zeros((num_points, 4))
 
         areas = (gt_bboxes[:, 2] - gt_bboxes[:, 0] + 1) * (
             gt_bboxes[:, 3] - gt_bboxes[:, 1] + 1)