diff --git a/answer_classifier_cached_features/inference.py b/answer_classifier_cached_features/inference.py
index 11082891125a6565fa6628dfa7a141ecc538a203..34b3e1dc47980822137e15d41e32a0bff717faab 100644
--- a/answer_classifier_cached_features/inference.py
+++ b/answer_classifier_cached_features/inference.py
@@ -8,6 +8,8 @@ class AnswerInference():
             self,
             object_feat,
             attribute_feat,
+            obj_detector_scores,
+            atr_detector_scores,
             answer_region_scores,
             question_vert_concat,
             answers_vert_concat,
@@ -44,11 +46,11 @@ class AnswerInference():
                     noun_embed.append(self.noun_embed[key1][j])
                     adjective_embed.append(self.adjective_embed[key2][j])
 
-                self.selected_noun_adjective[j] = self.inner_product_selection(
-                    self.object_feat[j], 
-                    self.attribute_feat[j],
-                    noun_embed,
-                    adjective_embed)
+                # self.selected_noun_adjective[j] = self.inner_product_selection(
+                #     self.object_feat[j], 
+                #     self.attribute_feat[j],
+                #     noun_embed,
+                #     adjective_embed)
             
 #            self.per_region_answer_scores = [None]*self.batch_size
             obj_atr_qa_feat = [None]*self.batch_size
@@ -76,9 +78,31 @@ class AnswerInference():
                     a_feat,
                     [self.num_regions, 1, 1])
 
+                # 100 x 1000
+                obj_det_feat = tf.expand_dims(
+                    self.obj_detector_scores[j],
+                    1)
+
+                # 100 x 18 x 1000
+                obj_det_feat = tf.tile(
+                    self.obj_det_feat,
+                    [1, self.num_answers, 1])
+
+                # 100 x 1000
+                atr_det_feat = tf.expand_dims(
+                    self.atr_detector_scores[j],
+                    1)
+
+                # 100 x 18 x 1000
+                atr_det_feat = tf.tile(
+                    self.atr_det_feat,
+                    [1, self.num_answers, 1])
+
                 obj_atr_qa_feat[j] = tf.concat(
                     2,
-                    [self.selected_noun_adjective[j], q_feat, a_feat])
+                    [obj_det_feat, atr_det_feat, q_feat, a_feat])
+
+                    # [self.selected_noun_adjective[j], q_feat, a_feat])
 
                 # obj_atr_qa_feat[j] = tf.expand_dims(
                 #     obj_atr_qa_feat[j],
diff --git a/answer_classifier_cached_features/train.py b/answer_classifier_cached_features/train.py
index 7d485fdb351bb0e69c2fc617b5988b17f8f8b425..294080382d739a612d6453b6d2d03591a1fed495 100644
--- a/answer_classifier_cached_features/train.py
+++ b/answer_classifier_cached_features/train.py
@@ -71,9 +71,19 @@ class graph_creator():
                 self.split_obj_atr_inference_output()
                 self.object_feat = self.object_embed_with_answers
                 self.attribute_feat = self.attribute_embed_with_answers
+                self.obj_detector_scores = self.object_scores_with_answers
+                self.atr_detector_scores = self.attribute_scores_with_answers
             else:
                 self.object_feat = self.obj_atr_inference.object_embed
                 self.attribute_feat = self.obj_atr_inference.attribute_embed
+                self.obj_detector_scores = tf.split(
+                    0,
+                    self.batch_size,
+                    self.obj_atr_inference.object_scores)
+                self.atr_detector_scores = tf.split(
+                    0,
+                    self.batch_size,
+                    self.obj_atr_inference.attribute_scores)
 
             self.object_feat = tf.split(
                 0,
@@ -105,6 +115,8 @@ class graph_creator():
             self.answer_inference = answer_graph.AnswerInference(
                 self.object_feat,
                 self.attribute_feat,
+                self.obj_detector_scores,
+                self.atr_detector_scores,
                 self.relevance_inference.answer_region_prob,
                 self.question_embed_concat,
                 self.answers_embed_concat,