</> Code with Papers by Syntology

Papers › 2nd Place Solution for Waymo Open Dataset Challenge - Real-time 2D Object Detection

2nd Place Solution for Waymo Open Dataset Challenge - Real-time 2D Object Detection

16 Jun 2021arXiv:2106.08713archive 2025-07-28

Yueming Zhang, Xiaolin Song, Bing Bai, Tengfei Xing, Chao Liu, Xin Gao, Zhihui Wang, Yawei Wen, Haojin Liao, Guoshan Zhang, Pengfei Xu

In an autonomous driving system, it is essential to recognize vehicles, pedestrians and cyclists from images. Besides the high accuracy of the prediction, the requirement of real-time running brings new challenges for convolutional network models. In this report, we introduce a real-time method to detect the 2D objects from images. We aggregate several popular one-stage object detectors and train the models of variety input strategies independently, to yield better performance for accurate multi-scale detection of each category, especially for small objects. For model acceleration, we leverage TensorRT to optimize the inference time of our detection pipeline. As shown in the leaderboard, our proposed detection framework ranks the 2nd place with 75.00% L1 mAP and 69.72% L2 mAP in the real-time 2D detection track of the Waymo Open Dataset Challenges, while our framework achieves the latency of 45.8ms/frame on an Nvidia Tesla V100 GPU.

PaperPDFCode

Code

WongKinYiu/yolor pytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

2D Object DetectionAutonomous DrivingObject Detectionobject-detection

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Object Detection Waymo Open Dataset LeapMotor_Det AP/L2 70.41 #1 of 5 Archive leaderboard report
Object Detection Waymo Open Dataset LeapMotor_Det Latency, ms 6.16 #1 of 5 Archive leaderboard report
Object Detection Waymo Open Dataset YOLOR_TensorRT (Ours) AP/L2 69.72 #2 of 5 Archive leaderboard report
Object Detection Waymo Open Dataset YOLOR_TensorRT (Ours) Latency, ms 4.58 #2 of 5 Archive leaderboard report
Object Detection Waymo Open Dataset YOLOR_P6_TRT AP/L2 69.56 #3 of 5 Archive leaderboard report
Object Detection Waymo Open Dataset YOLOR_P6_TRT Latency, ms 3.74 #3 of 5 Archive leaderboard report
Object Detection Waymo Open Dataset dereyly_self_ensemble AP/L2 65.65 #4 of 5 Archive leaderboard report
Object Detection Waymo Open Dataset dereyly_self_ensemble Latency, ms 6.87 #4 of 5 Archive leaderboard report
Object Detection Waymo Open Dataset YOLO_v5 AP/L2 64.14 #5 of 5 Archive leaderboard report
Object Detection Waymo Open Dataset YOLO_v5 Latency, ms 3.81 #5 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections