Related Objects of Interest: car , traffic signal , truck , person , bus , stop , bicycle , bike , no entry , roundabout

Showing 1 - 50 of 106

1 2 3

Top Traffic Signal Computer Vision Models

The models below have been fine-tuned for various traffic signal detection tasks. You can try out each model in your browser, or test an edge deployment solution (i.e. to an NVIDIA Jetson). You can use the datasets associated with the models below as a starting point for building your own traffic signal detection model.

At the bottom of this page, we have guides on how to count traffic signals in images and videos.

Object Detection Model

Object detection

Vishi Sapra

900 images1 model

bike bus car person traffic signal truck

Object Detection Model

Object Detection

risproject

1822 images1 model

Duckie Duckiebot Intersection sign QR code Signal sign Stop sign Traffic light

Object Detection Model

10000 images2 models

bike biker bus car crosswalk human motor motorbike pedestrian person rider traffic light traffic sign traffic signal truck van 0 1 2 3

Object Detection Model snap

pedestrian Traffic Light

ono

926 images3 models

green red trafficlight pedestrian Traffic Light signal-light traffic_light

Object Detection Model

trafficvideo

636 images1 model

bicycle car pedestrian traffic signal truck van

Object Detection Model

vehicle_detection

NED University of engineering and technology

41 images1 model

bicycle bus car people traffic signal truck peple

Object Detection Model

traffic sign

python

704 images1 model

green green light parking red red light stop left right traffic-signal yellow light

Object Detection Model

Intelligent Vehicle Traffic Recognition

National University of Singapore

968 images1 model

bench bus car fire hydrant person stop sign traffic light traffic signal truck

Object Detection Model

Car Object Detection

Parking Project

1137 images1 model

bus car person traffic signal truck Car Empty

Object Detection Model

Signal

SRHBerlin

52 images2 models

Traffic-Signal

Object Detection Model yolov8s yolov8 snap

rail-road-crossing

AL AMIN

1119 images4 models

car person train truck barrier barrier-closed rails traffic-signal

Object Detection Model

paper-yolo

dats2022

2672 images1 model

cow person Ambulance Barricade Bike Board Bridge Building Bus Bus Stop Car Cart Cattle Crane Cycle Dog Electricity Pole Flag Goat Lamp Post

Object Detection Model

Combined Dataset Car Detection

Parking Project

1232 images1 model

bus car person traffic signal truck Car Empty

Object Detection Model

main project

RV College of Engineering

572 images3 models

Car Moter bike Red light Traffic Signal

Object Detection Model

AutoDrive Deer Data

15fram@queensu.ca

620 images1 model

animal deer pedestrian fixed-object-barricade four-wheel-vehicle large-vehicle no-parking one-way-r other-traffic-signal speed-40-km stop-sign three-light-traffic

Object Detection Model

w251-homework-8

kkasravi

161 images1 model

bus car person sign traffic signal truck cross walk street lamp

Object Detection Model

MIYO

midasyolo

300 images3 models

bus car pedestrian traffic signal truck traffic symbol

Object Detection Model

rick vick

898 images1 model

bus car person traffic signal truck

Object Detection Model snap

Road Object detection

Intel Project

3453 images1 model

bicycle bridge bus car motorcycle person road sign traffic signal train truck autorickshaw car'streetlight

Object Detection Model snap

Image dataset

Pradeep

855 images1 model

car chair 0 Highway Start Highway end No entry Parking Pedestrian Crossing Roundabout Stop Straight Toy priority signal traffic-signs

Object Detection Model

Vehicle Detection

Power House

4720 images2 models

car dog Animal Bill Board Building Car Cats Door House Motorcycle Pavement People Person Pole Sign board Stone Street Light Traffic signal Tree Wall

Object Detection Model snap

custom_data

Project1

4857 images1 model

apple boat bottle bus car cellphone chair clock fire fork gate hammer handle knife lorry manhole oven pen scissors ship

Semantic Segmentation Model

My First Project

GemDetect

109 images1 model

auto bicycle bike building bus car cart kite motorcycle person pole potted plant road scooter sidewalk tower traffic light train tree truck

Object Detection Model

Deep Learning - Group Assignment

4869 images1 model

0 2-wheeler 0 2-wheeler 1 auto 1 auto 2 bus 2 bus 3 car 3 car 4 crosswalk 5 person 6 traffic signal 7 truck 7 truck 8 twowheeler 9 van

Object Detection Model

10000 images2 models

bike biker bus car crosswalk human motor motorbike pedestrian person rider traffic light traffic sign traffic signal truck van 0 1 2 3

Object Detection Model

Airsim

Satyam

65 images1 model

Person cars cars-people-traffic-signal-

Classification Model

Fpoly

9476 images1 model

bike bus parking pedestrian red road sign stop (reverse 15 20 25 30 35 40 45 50 55 60 65

Object Detection Model

mde

Mde

300 images1 model

car pedestrian scooter traffic signal truck traffic symbol

Object Detection Model snap

car models

gmu

898 images1 model

bus car person traffic signal truck

Object Detection Model snap

vehicles

eastsky

100 images1 model

bus car person traffic signal truck

Object Detection Model snap

Signal Light Detection

hihello

1802 images1 model

airplane apple backpack banana baseball bat bench bicycle bird boat book bottle bus car cat cell phone chair clock couch cow cup

Object Detection Model snap

pole facilities

Beijing University of Technology

165 images1 model

base-station-pole lighting-pole predestrain-guidence-signage-pole surveillance-pole traffic-sign-pole traffic-signal-light-pole tramway-pole

Showing 1 - 50 of 106

1 2 3

Guide: How to Count Traffic Signals with Computer Vision

With a model hosted on Roboflow like the ones above and the open source supervision Python package, you can count traffic signals in your images and videos.

The following code snippet counts the number of traffic signals present in an image.

To use the snippet below, you will need to run pip install roboflow supervision. Replace the project name and model name with any model trained on Universe, such as those listed above.

import supervision as sv
            import roboflow
            
            roboflow.login()
            rf = roboflow.Roboflow()
            
            # replace with the traffic signal project you choose above
            project = rf.workspace("vishi-sapra").project("object-detection-axukj")
            traffic_signal_model = project.version(10).model
            
            results = traffic_signal_model.predict("traffic_signal.jpg").json()
            traffic_signals = sv.Detections.from_roboflow(results)
            
            # print number of traffic signals
            print(len(traffic_signals))

Guide: How to Count Traffic Signals in a Zone

With a bit more code, you can count the number of traffic signal present in a specific zone of your image or video.

The following code snippet counts the number of traffic signal present in each frame in a video.

To use the snippet below, you will need to run pip install roboflow supervision. Replace the project name and model name with any model trained on Universe, such as those listed above.

Read our blog post on counting objects in a zone

import numpy as np
            import supervision as sv
            import roboflow
            
            SOURCE_VIDEO_PATH = "traffic_signal.mp4"
            TARGET_VIDEO_PATH = "traffic_signal_out.mp4"
            
            # use https://roboflow.github.io/polygonzone/ to get the points for your shape
            polygon = np.array([
                # draw 50x50 box in top left corner
                [0, 0],
                [50, 0],
                [50, 50],
                [0, 50]
            ])
            
            roboflow.login()
            rf = roboflow.Roboflow()
            
            # replace with the traffic signal project you choose above
            project = rf.workspace("vishi-sapra").project("object-detection-axukj")
            traffic_signal_model = project.version(10).model
            
            # create BYTETracker instance
            traffic_signal_tracker = sv.ByteTrack(track_thresh=0.25, track_buffer=30, match_thresh=0.8, frame_rate=30)
            
            # create VideoInfo instance
            video_info = sv.VideoInfo.from_video_path(SOURCE_VIDEO_PATH)
            
            # create frame generator
            generator = sv.get_video_frames_generator(SOURCE_VIDEO_PATH)
            
            # create PolygonZone instance
            zone = sv.PolygonZone(polygon=polygon, frame_resolution_wh=(video_info.width, video_info.height))
            
            # create box annotator
            box_annotator = sv.BoxAnnotator(thickness=4, text_thickness=4, text_scale=2)
            
            colors = sv.ColorPalette.default()
            
            # create instance of BoxAnnotator
            zone_annotator = sv.PolygonZoneAnnotator(thickness=4, text_thickness=4, text_scale=2, zone=zone, color=colors.colors[0])
            
            # define call back function to be used in video processing
            def callback(frame: np.ndarray, index:int) -> np.ndarray:
                # model prediction on single frame and conversion to supervision Detections
                results = traffic_signal_model.predict(frame).json()
                traffic_signals = sv.Detections.from_roboflow(results)
            
                # show traffic signal detections in real time
                print(traffic_signals)
            
                # tracking traffic signal detections
                traffic_signals = traffic_signal_tracker.update_with_detections(traffic_signals)
            
                annotated_frame = box_annotator.annotate(scene=frame, detections=traffic_signals)
                annotated_frame = zone_annotator.annotate(scene=annotated_frame)
            
                # return frame with box and line annotated result
                return annotated_frame
            
            # process the whole video
            sv.process_video(
                source_path = SOURCE_VIDEO_PATH,
                target_path = TARGET_VIDEO_PATH,
                callback=callback
            )

Guide: How to Track Traffic Signals Crossing a Line

You can count how many traffic signals have crossed a line using the supervision LineCounter method.

The following code snippet counts the number of traffic signals that cross a line in a video.

To use the snippet below, you will need to run pip install roboflow supervision. Replace the project name and model name with any model trained on Universe, such as those listed above.

import numpy as np
            import supervision as sv
            import roboflow
            
            SOURCE_VIDEO_PATH = "traffic_signal.mp4"
            TARGET_VIDEO_PATH = "traffic_signal_out.mp4"
            
            # use https://roboflow.github.io/polygonzone/ to get the points for your line
            LINE_START = sv.Point(0, 300)
            LINE_END = sv.Point(800, 300)
            
            roboflow.login()
            rf = roboflow.Roboflow()
            
            # replace with the traffic signal project you choose above
            project = rf.workspace("vishi-sapra").project("object-detection-axukj")
            traffic_signal_model = project.version(10).model
            
            # create BYTETracker instance
            traffic_signal_tracker = sv.ByteTrack(track_thresh=0.25, track_buffer=30, match_thresh=0.8, frame_rate=30)
            
            # create VideoInfo instance
            video_info = sv.VideoInfo.from_video_path(SOURCE_VIDEO_PATH)
            
            # create frame generator
            generator = sv.get_video_frames_generator(SOURCE_VIDEO_PATH)
            
            # create LineZone instance, it is previously called LineCounter class
            line_zone = sv.LineZone(start=LINE_START, end=LINE_END)
            
            # create instance of BoxAnnotator
            box_annotator = sv.BoxAnnotator(thickness=4, text_thickness=4, text_scale=2)
            
            # create instance of TraceAnnotator
            trace_annotator = sv.TraceAnnotator(thickness=4, trace_length=50)
            line_zone_annotator = sv.LineZoneAnnotator(thickness=4, text_thickness=4, text_scale=2)
            
            # define call back function to be used in video processing
            def callback(frame: np.ndarray, index:int) -> np.ndarray:
                # model prediction on single frame and conversion to supervision Detections
                results = traffic_signal_model.predict(frame).json()
                traffic_signals = sv.Detections.from_roboflow(results)
            
                # show traffic signal detections in real time
                print(traffic_signals)
            
                # tracking traffic signal detections
                traffic_signals = traffic_signal_tracker.update_with_detections(traffic_signals)
                annotated_frame = trace_annotator.annotate(
                    scene=frame.copy(),
                    detections=traffic_signals
                )
                annotated_frame=box_annotator.annotate(
                    scene=annotated_frame,
                    detections=traffic_signals
                )
            
                # update line counter
                line_zone.trigger(traffic_signals)
            
                # return frame with box and line annotated result
                return line_zone_annotator.annotate(annotated_frame, line_counter=line_zone)
            
            # process the whole video
            sv.process_video(
                source_path = SOURCE_VIDEO_PATH,
                target_path = TARGET_VIDEO_PATH,
                callback=callback
            )