OpenSurfaces

By Sean Bell, Paul Upchurch, Noah Snavely, Kavita Bala. 2013

145 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, Geology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

In Geology

Citations in the corpus

ImageNet Large Scale Visual Recognition Challenge (2014, background): “ images annotated with object boundaries. OpenSurfaces segments surfaces from consumer photographs and annotates them with surface properties, including material, texture, and contextual information (Bell et al., 2013) . The closest to ILSVRC is the PASCAL VOC dataset (Everingham et al., 2010, 2014)”
Semantic Understanding of Scenes Through the ADE20K Dataset (2016, methodology): “forts include Pascal-Context [22], NYU Depth V2 [23], SUN database [34], SUN RGB-D dataset [31], CityScapes Semantic Understanding of Scenes through the ADE20K Dataset 3 dataset [7], and OpenSurfaces [2, 3]. Recently COCO stuff dataset [4] provides additional stuff segmentation complementary to the ”
Interpretability Beyond Classification Output: Semantic Bottleneck Networks (2019, methodology): “Based on the 377 part and material concepts available (351 parts sourced from ADE [32] and Pascal-Part [7] and 26 materials sourced from OpenSurfaces [5]), we compile multiple subsets containing at least 6 concepts in order to explore the relationship between number of concepts and segmentation perf”
BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning (2018, methodology): “uation benchmark for computer vision research for autonomous driving. 2 Related Works Labeling tools have played an important role in generating annotations for supervised learning in computer vision [18,14,20,3]. Russell et al. [18] introduced a web-based BDD100K 3 labeling tool, LabelMe, which is ”
ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentation (2016, methodology): “For example, powerful interactive tools [1] are adopted for annotating the MS COCO dataset [18], but it still takes minutes for an experienced annotator labeling one image [18].”
Semantically-enriched 3D models for common-sense knowledge (2015, methodology): “As part of a larger pipeline we plan to augment these algorithmic predictions through manual annotation and verification by people using crowdsourcing.”
Attentive Explanations: Justifying Decisions and Pointing to the Evidence (2016, methodology): “ to where humans think the evidence for the answer is. Human-annotated attention maps are collected via Amazon Mechanical Turk where we use the segmentation UI interface from the OpenSurfaces Project [6]. Annotators are provided with an image and an answer (question and answer pair for VQA-X, class ”
Transient attributes for high-level understanding and editing of outdoor scenes (2014, methodology): “For example, crowdsourcing has enabled object insertion in Photo Clip Art [Lalonde et al. 2007], analysis and recognition of human object sketches [Eitz et al. 2012], and extensive surface appearance reference data in [Bell et al. 2013].”
Neural Inverse Rendering of an Indoor Scene From a Single Image (2019, methodology): “, pair-wise reflectance comparison [2, 19, 47], sparse material segmentation [3]) as a form of weakly supervised learning, to further improve the performance on real images.”
Unite the People: Closing the Loop Between 3D and 2D Human Representations (2017, methodology): “We built an interactive annotation tool on top of the Opensurfaces package [3] to work with Amazon Mechanical Turk (AMT).”
From BoW to CNN: Two Decades of Texture Representation for Texture Classification (2018, methodology): “Each material instance was measured by a full bidirectional texture function with Subsequent to FMD, Bell et al. [13] released OpenSurfaces (OS) which has over 20,000 images from consumer photographs, each containing a number of high-quality texture or material segments.”
Semantic Amodal Segmentation (2015, methodology): “Figure 3: A screenshot of our annotation tool for semantic amodal segmentation (adopted from the Open Surfaces tool [3]).”
Understanding RealWorld Indoor Scenes with Synthetic Data (2015, unknown): “…that basis scenes will be useful as a standalone dataset for tasks such as benchmarking the performance within Simultaneous Localisation and Mapping in robotics (SLAM) [21] as well as providing environments for reinforcement learning and physical scene understanding with a physics engine [34].”
In the Eye of the Beholder: Gaze and Actions in First Person Video (2020, methodology): “modified interface from [97] was used to contour the egocentric hands.”
Deep Filter Banks for Texture Recognition, Description, and Segmentation (2015, methodology): “Rather than collecting a new image dataset from scratch, our starting point is the excellent open surfaces (OS) dataset that was recently introduced by Bell et al. (2013).”