DDD17: End-To-End DAVIS Driving Dataset

Binas, Jonathan; Neil, Daniel; Liu, Shih-Chii; Delbrück, Tobi

doi:10.48550/arxiv.1711.01458

Cited by 19 publications

(29 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DDD17 [60] contains 40 different driving sequences of event data captured by a DVS camera. While the dataset provides both grayscale images and event data, it does not provide semantic segmentation labels.…”

Section: Experiments a Experimental Setup 1) Datasetmentioning

confidence: 99%

Beyond Classification: Directly Training Spiking Neural Networks for Semantic Segmentation

Kim¹,

Chough²,

Panda³

2021

Preprint

View full text Add to dashboard Cite

Spiking Neural Networks (SNNs) have recently emerged as the low-power alternative to Artificial Neural Networks (ANNs) because of their sparse, asynchronous, and binary event-driven processing. Due to their energy efficiency, SNNs have a high possibility of being deployed for real-world, resourceconstrained systems such as autonomous vehicles and drones. However, owing to their non-differentiable and complex neuronal dynamics, most previous SNN optimization methods have been limited to image recognition. In this paper, we explore the SNN applications beyond classification and present semantic segmentation networks configured with spiking neurons. Specifically, we first investigate two representative SNN optimization techniques for recognition tasks (i.e., ANN-SNN conversion and surrogate gradient learning) on semantic segmentation datasets. We observe that, when converted from ANNs, SNNs suffer from high latency and low performance due to the spatial variance of features. Therefore, we directly train networks with surrogate gradient learning, resulting in lower latency and higher performance than ANN-SNN conversion. Moreover, we redesign two fundamental ANN segmentation architectures (i.e., Fully Convolutional Networks and DeepLab) for the SNN domain. We conduct experiments on two public semantic segmentation benchmarks including the PASCAL VOC2012 dataset and the DDD17 event-based dataset. In addition to showing the feasibility of SNNs for semantic segmentation, we show that SNNs can be more robust and energy-efficient compared to their ANN counterparts in this domain.

show abstract

Section: Experiments a Experimental Setup 1) Datasetmentioning

confidence: 99%

Beyond Classification: Directly Training Spiking Neural Networks for Semantic Segmentation

Kim¹,

Chough²,

Panda³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…According to the image-like transformation, Alonso et al [57] introduced a six-channels event representation and constructed a semantic segmentation model Ev-SegNet on an extended event dataset DDD17 [58], whose semantic labels are generated by a pre-trained model on Cityscapes and only contain 6 major categories. In contrast, our models are trained with the ground-truth labels of Cityscapes and perform semantic segmentation in all 19 object classes, so that the perception component can deliver a sufficiently dense and finegrained scene understanding result for upper-level assistance systems.…”

Section: Event-based Visionmentioning

confidence: 99%

Exploring Event-driven Dynamic Context for Accident Scene Segmentation

Zhang¹,

Yang²,

Stiefelhagen³

2021

Preprint

View full text Add to dashboard Cite

The robustness of semantic segmentation on edge cases of traffic scene is a vital factor for the safety of intelligent transportation. However, most of the critical scenes of traffic accidents are extremely dynamic and previously unseen, which seriously harm the performance of semantic segmentation methods. In addition, the delay of the traditional camera during highspeed driving will further reduce the contextual information in the time dimension. Therefore, we propose to extract dynamic context from event-based data with a higher temporal resolution to enhance static RGB images, even for those from traffic accidents with motion blur, collisions, deformations, overturns, etc. Moreover, in order to evaluate the segmentation performance in traffic accidents, we provide a pixel-wise annotated accident dataset, namely DADA-seg, which contains a variety of critical scenarios from traffic accidents. Our experiments indicate that event-based data can provide complementary information to stabilize semantic segmentation under adverse conditions by preserving fine-grained motion of fast-moving foreground (crash objects) in accidents. Our approach achieves +8.2% performance gain on the proposed accident dataset, exceeding more than 20 state-of-the-art semantic segmentation methods. The proposal has been demonstrated to be consistently effective for models learned on multiple source databases including Cityscapes, KITTI-360, BDD, and ApolloScape.

show abstract

“…In [5], 12 hours of driving sequences are obtained during day and night time. Various car information, such as vehicle speed, GPS position, driver steering angle, are associated to the dataset.…”

Section: Event-based Datasets For Visual Odometry Optical Flow and St...mentioning

confidence: 99%

“…As shown by Tab. 1, the GEN1 Automotive Detection Dataset is 3 time larger than the DDD17 [5] dataset in terms of hours and has about 22 times more labels than the [31] pedestrian dataset. In terms of number of labels, the [4] dataset is the second largest one, with approximately 2.5 less labels than ours.…”

Section: Analysis and Statisticsmentioning

confidence: 99%

A Large Scale Event-based Detection Dataset for Automotive

Tournemire¹,

Nitti²,

Pérot³

et al. 2020

Preprint

View full text Add to dashboard Cite

We introduce the first very large detection dataset for event cameras. The dataset is composed of more than 39 hours of automotive recordings acquired with a 304x240 GEN1 sensor. It contains open roads and very diverse driving scenarios, ranging from urban, highway, suburbs and countryside scenes, as well as different weather and illumination conditions.Manual bounding box annotations of cars and pedestrians contained in the recordings are also provided at a frequency between 1 and 4Hz, yielding more than 255,000 labels in total. We believe that the availability of a labeled dataset of this size will contribute to major advances in event-based vision tasks such as object detection and classification. We also expect benefits in other tasks such as optical flow, structure from motion and tracking, where for example, the large amount of data can be leveraged by selfsupervised learning methods.

show abstract

DDD17: End-To-End DAVIS Driving Dataset

Cited by 19 publications

References 7 publications

Beyond Classification: Directly Training Spiking Neural Networks for Semantic Segmentation

Beyond Classification: Directly Training Spiking Neural Networks for Semantic Segmentation

Exploring Event-driven Dynamic Context for Accident Scene Segmentation

A Large Scale Event-based Detection Dataset for Automotive

Contact Info

Product

Resources

About