fix default settings for demo

README.md

@@ -69,15 +69,15 @@ Step1. Download a pretrained model from the benchmark table.
 Step2. Use either -n or -f to specify your detector's config. For example:
 
 ```shell
-python tools/demo.py image -n yolox-s -c /path/to/your/yolox_s.pth.tar --path assets/dog.jpg --conf 0.3 --nms 0.65 --tsize 640 --save_result --device [cpu/gpu]
+python tools/demo.py image -n yolox-s -c /path/to/your/yolox_s.pth.tar --path assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
 ```
 or
 ```shell
-python tools/demo.py image -f exps/default/yolox_s.py -c /path/to/your/yolox_s.pth.tar --path assets/dog.jpg --conf 0.3 --nms 0.65 --tsize 640 --save_result --device [cpu/gpu]
+python tools/demo.py image -f exps/default/yolox_s.py -c /path/to/your/yolox_s.pth.tar --path assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
 ```
 Demo for video:
 ```shell
-python tools/demo.py video -n yolox-s -c /path/to/your/yolox_s.pth.tar --path /path/to/your/video --conf 0.3 --nms 0.65 --tsize 640 --save_result --device [cpu/gpu]
+python tools/demo.py video -n yolox-s -c /path/to/your/yolox_s.pth.tar --path /path/to/your/video --conf 0.25 --nms 0.45 --tsize 640 --save_result --device [cpu/gpu]
 ```
 
 

demo/ONNXRuntime/onnx_inference.py

@@ -83,8 +83,7 @@ if __name__ == '__main__':
     boxes_xyxy[:, 2] = boxes[:, 0] + boxes[:, 2]/2.
     boxes_xyxy[:, 3] = boxes[:, 1] + boxes[:, 3]/2.
     boxes_xyxy /= ratio
-    dets = multiclass_nms(boxes_xyxy, scores, nms_thr=0.65, score_thr=0.1)
-    
+    dets = multiclass_nms(boxes_xyxy, scores, nms_thr=0.45, score_thr=0.1)
     if dets is not None:
         final_boxes, final_scores, final_cls_inds = dets[:, :4], dets[:, 4], dets[:, 5]
         origin_img = vis(origin_img, final_boxes, final_scores, final_cls_inds,

demo/OpenVINO/cpp/yolox_openvino.cpp

@@ -18,7 +18,7 @@ using namespace InferenceEngine;
 #define tcout                  std::cout
 #define file_name_t            std::string
 #define imread_t               cv::imread
-#define NMS_THRESH 0.65
+#define NMS_THRESH 0.45
 #define BBOX_CONF_THRESH 0.3
 
 static const int INPUT_W = 416;

demo/OpenVINO/python/openvino_inference.py

@@ -141,8 +141,8 @@ def main():
     boxes_xyxy[:, 2] = boxes[:, 0] + boxes[:, 2]/2.
     boxes_xyxy[:, 3] = boxes[:, 1] + boxes[:, 3]/2.
     boxes_xyxy /= ratio
-    dets = multiclass_nms(boxes_xyxy, scores, nms_thr=0.65, score_thr=0.1)
-    
+    dets = multiclass_nms(boxes_xyxy, scores, nms_thr=0.45, score_thr=0.1)
+
     if dets is not None:
         final_boxes = dets[:, :4]
         final_scores, final_cls_inds = dets[:, 4], dets[:, 5]

demo/TensorRT/cpp/yolox.cpp

@@ -22,7 +22,7 @@
     } while (0)
 
 #define DEVICE 0  // GPU id
-#define NMS_THRESH 0.65
+#define NMS_THRESH 0.45
 #define BBOX_CONF_THRESH 0.3
 
 using namespace nvinfer1;

tools/demo.py

@@ -81,12 +81,13 @@ def get_image_list(path):
 
 
 class Predictor(object):
-    def __init__(self, model, exp, cls_names=COCO_CLASSES, trt_file=None, decoder=None, device="cpu"):
+    def __init__(self, model, exp, cls_names=COCO_CLASSES, trt_file=None, decoder=None, conf_vis=0.3, device="cpu"):
         self.model = model
         self.cls_names = cls_names
         self.decoder = decoder
         self.num_classes = exp.num_classes
         self.confthre = exp.test_conf
+        self.conv_vis = conf_vis
         self.nmsthre = exp.nmsthre
         self.test_size = exp.test_size
         self.device = device
@@ -158,7 +159,7 @@ def image_demo(predictor, vis_folder, path, current_time, save_result):
     files.sort()
     for image_name in files:
         outputs, img_info = predictor.inference(image_name)
-        result_image = predictor.visual(outputs[0], img_info)
+        result_image = predictor.visual(outputs[0], img_info, predictor.conf_vis)
         if save_result:
             save_folder = os.path.join(
                 vis_folder, time.strftime("%Y_%m_%d_%H_%M_%S", current_time)
@@ -191,7 +192,7 @@ def imageflow_demo(predictor, vis_folder, current_time, args):
         ret_val, frame = cap.read()
         if ret_val:
             outputs, img_info = predictor.inference(frame)
-            result_frame = predictor.visual(outputs[0], img_info)
+            result_frame = predictor.visual(outputs[0], img_info, predictor.conf_vis)
             if args.save_result:
                 vid_writer.write(result_frame)
             ch = cv2.waitKey(1)
@@ -260,7 +261,7 @@ def main(exp, args):
         trt_file = None
         decoder = None
 
-    predictor = Predictor(model, exp, COCO_CLASSES, trt_file, decoder, args.device)
+    predictor = Predictor(model, exp, COCO_CLASSES, trt_file, decoder, args.conf, args.device)
     current_time = time.localtime()
     if args.demo == 'image':
         image_demo(predictor, vis_folder, args.path, current_time, args.save_result)