diff --git a/docs/specs/images/Apollo3.5_perception_detail.png b/docs/specs/images/Apollo3.5_perception_detail.png
new file mode 100644
index 0000000000000000000000000000000000000000..7698e1952a9c4f5858d834baeb8bbda5d143095b
Binary files /dev/null and b/docs/specs/images/Apollo3.5_perception_detail.png differ
diff --git a/docs/specs/images/Apollo3.5_perception_sensor_based.png b/docs/specs/images/Apollo3.5_perception_sensor_based.png
new file mode 100644
index 0000000000000000000000000000000000000000..02be3b608e7e843ed9ae37f73a01de71d6d00665
Binary files /dev/null and b/docs/specs/images/Apollo3.5_perception_sensor_based.png differ
diff --git a/modules/perception/README.md b/modules/perception/README.md
index 6b1a240edc6255d2ed4b2525f653b6b46b3c502b..0c0fd32a0145bf0836211402dc3cf2542e7ea26a 100644
--- a/modules/perception/README.md
+++ b/modules/perception/README.md
@@ -12,6 +12,13 @@ Apollo Perception has following new features:
 
 The perception module incorporates the capability of using 5 cameras (2 front, 2 on either side and 1 rear) and 2 radars (front and rear) along with 3 16-line LiDARs (2 rear and 1 front) and 1 128-line LiDAR to recognize obstacles and fuse their individual tracks to obtain a final track list. The obstacle sub-module detects, classifies and tracks obstacles. This sub-module also predicts obstacle motion and position information (e.g., heading and velocity). For lane line, we construct lane instances by postprocessing lane parsing pixels and calculate the lane relative location to the ego-vehicle (L0, L1, R0, R1, etc.).
 
+## Architecture
+
+The general architecture of the perception module is shown:
+![](docs/specs/images/Apollo3.5_perception_sensor_based.png)
+
+The detailed perception modules are displayed below.
+![](docs/specs/images/Apollo3.5_perception_detail.png)
 
 ## Input
 
@@ -29,8 +36,8 @@ The perception module inputs are:
 
 The perception module outputs are:
 
-* The 3D obstacle tracks with the heading, velocity and classification information (ROS topic /apollo/perception/obstacles_)
-* The lane marker information with fitted curve parameter, spatial information(L0, R0, etc.) as well as semantic information (lane type) (cyber channel /apollo/perception/obstacles_)
+* The 3D obstacle tracks with the heading, velocity and classification information (cyber channel /apollo/perception/obstacles)
+* The output of traffic light detection and recognition (cyber channel /apollo/perception/traffic_light)
 
 ## Setup Instructions