diff --git a/lite/kernels/host/matrix_nms_compute.cc b/lite/kernels/host/matrix_nms_compute.cc
index b5c381e3f91df2cb20bf8317f3b430c373b42567..2fe4e6f9723a6bd5933f3c6638b9492c0cdac137 100644
--- a/lite/kernels/host/matrix_nms_compute.cc
+++ b/lite/kernels/host/matrix_nms_compute.cc
@@ -273,8 +273,8 @@ void MatrixNmsCompute::Run() {
   auto out_dim = box_dim + 2;
 
   Tensor boxes_slice, scores_slice;
-  size_t num_out = 0;
-  std::vector<uint64_t> offsets = {0};
+  int64_t num_out = 0;
+  std::vector<int64_t> offsets = {0};
   std::vector<float> detections;
   std::vector<int> indices;
   detections.reserve(out_dim * num_boxes * batch_size);
@@ -298,23 +298,25 @@ void MatrixNmsCompute::Run() {
                                   post_threshold,
                                   use_gaussian,
                                   gaussian_sigma);
-    offsets.push_back(offsets.back() + static_cast<uint64_t>(num_out));
+    offsets.push_back(offsets.back() + num_out);
   }
 
-  uint64_t num_kept = offsets.back();
+  int64_t num_kept = offsets.back();
   if (num_kept == 0) {
     outs->Resize({0, out_dim});
+    outs->mutable_data<float>();
     index->Resize({0, 1});
+    index->mutable_data<int>();
   } else {
-    outs->Resize({static_cast<int64_t>(num_kept), out_dim});
-    index->Resize({static_cast<int64_t>(num_kept), 1});
+    outs->Resize({num_kept, out_dim});
+    index->Resize({num_kept, 1});
     std::copy(
         detections.begin(), detections.end(), outs->mutable_data<float>());
     std::copy(indices.begin(), indices.end(), index->mutable_data<int>());
   }
 
   LoD lod;
-  lod.emplace_back(offsets);
+  lod.emplace_back(std::vector<uint64_t>(offsets.begin(), offsets.end()));
   outs->set_lod(lod);
   index->set_lod(lod);
 }