diff --git a/mmdet/models/dense_heads/paa_head.py b/mmdet/models/dense_heads/paa_head.py
index 85a80e844..90fad1147 100644
--- a/mmdet/models/dense_heads/paa_head.py
+++ b/mmdet/models/dense_heads/paa_head.py
@@ -120,7 +120,6 @@ class PAAHead(ATSSHead):
         bbox_preds = [item.reshape(-1, 4) for item in bbox_preds]
         iou_preds = levels_to_images(iou_preds)
         iou_preds = [item.reshape(-1, 1) for item in iou_preds]
-
         pos_losses_list, = multi_apply(self.get_pos_loss, anchor_list,
                                        cls_scores, bbox_preds, labels,
                                        labels_weight, bboxes_target,
@@ -138,6 +137,8 @@ class PAAHead(ATSSHead):
                 anchor_list,
             )
             num_pos = sum(num_pos)
+            if num_pos == 0:
+                num_pos = len(img_metas)
         # convert all tensor list to a flatten tensor
         cls_scores = torch.cat(cls_scores, 0).view(-1, cls_scores[0].size(-1))
         bbox_preds = torch.cat(bbox_preds, 0).view(-1, bbox_preds[0].size(-1))
@@ -203,6 +204,8 @@ class PAAHead(ATSSHead):
         Returns:
             Tensor: Losses of all positive samples in single image.
         """
+        if not len(pos_inds):
+            return cls_score.new([]),
         anchors_all_level = torch.cat(anchors, 0)
         pos_scores = cls_score[pos_inds]
         pos_bbox_pred = bbox_pred[pos_inds]
@@ -359,6 +362,8 @@ class PAAHead(ATSSHead):
         # https://github.com/kkhoot/PAA/issues/8 and
         # https://github.com/kkhoot/PAA/issues/9.
         fgs = gmm_assignment == 0
+        pos_inds_temp = fgs.new_tensor([], dtype=torch.long)
+        ignore_inds_temp = fgs.new_tensor([], dtype=torch.long)
         if fgs.nonzero().numel():
             _, pos_thr_ind = scores[fgs].topk(1)
             pos_inds_temp = pos_inds_gmm[fgs][:pos_thr_ind + 1]