Intrinsic images in the wild

By Sean Bell, Kavita Bala, Noah Snavely. 2014

490 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding (2020, background): “Together, these features make our dataset well-suited for geometric learning problems that require direct 3D supervision (e.g., [27]), multi-task learning problems that require reasoning jointly over multiple input and output modalities (e.g., [71]), and inverse rendering problems (e.g., [13, 40]).”
Network Dissection: Quantifying Interpretability of Deep Visual Representations (2017, background): “The Broadly and Densely Labeled Dataset (Broden) unifies several densely labeled image data sets: ADE [43], OpenSurfaces [4], Pascal-Context [19], Pascal-Part [6], and the Describable Textures Dataset [7].”
Interpreting Deep Visual Representations via Network Dissection (2017, background): “scene 468 ADE [32] 38 object 584 ADE [32], Pascal-Context [34] 491 part 234 ADE [32], Pascal-Part [35] 854 material 32 OpenSurfaces [33] 1,703 texture 47 DTD [36] 140 color 11 Generated 59,250 Fig.”
Learning to Reconstruct Shapes from Unseen Classes (2018, background): “Through years, researchers have explored recovering 2.5D sketches from texture, shading, or color images [Barron and Malik, 2015, Bell et al., 2014, Horn and Brooks, 1989, Tappen et al., 2003, Weiss, 2001, Zhang et al., 1999].”
Learning Shape Priors for Single-View 3D Completion and Reconstruction (2018, background): “In the past, researchers have explored recovering 2.5D sketches from shading, texture, or color images [2,3,20,47,55,63].”
MarrNet: 3D Shape Reconstruction via 2.5D Sketches (2017, background): “In the past, researchers have explored recovering 2.5D shape from shading, texture, or color images [Horn and Brooks, 1989, Zhang et al., 1999, Tappen et al., 2003, Barron and Malik, 2015, Weiss, 2001, Bell et al., 2014].”
Matterport3D: Learning from RGB-D Data in Indoor Environments (2017, background): “…multiplicity and diversity of views enables opportunities for learning to predict view-dependent surface properties, such as material reﬂectance [4, 26], and for learning to factor out view-dependence when learning view-independent representations, such as patch descriptors [45, 46] and normals…”
Learning Non-Lambertian Object Intrinsics Across ShapeNet Categories (2016, methodology): “Our method is evaluated and compared with SIRFS [3], IIW [5], and Direct Intrinsics (DI) [20].”
SfSNet: Learning Shape, Reflectance and Illuminance of Faces 'in the Wild' (2017, background): “Although our work concentrates on decomposing faces, the problem of inverse rendering for generic objects in a learning based framework has also gained attention in recent years [4, 20, 29, 11].”
Intrinsic Decompositions for Image Editing (2017, unknown): “Recent datasets like IIW [BBS14] have expanded the evaluation to more real world setting using photographs, showing that more progress is required.”
Ordinal Depth Supervision for 3D Human Pose Estimation (2018, unknown): “[3] and Chen et al .”
CGIntrinsics: Better Intrinsic Image Decomposition through Physically-Based Rendering (2018, methodology): “MPI Sintel [34] 890 Animation non-PB spatial-varying full MIT Intrinsics [35] 110 Object Real single global full ShapeNet [2] 2M+ Object PB single global full IIW [5] 5230 Scene Real spatial-varying sparse SAW [6] 6677 Scene Real spatial-varying sparse CGINTRINSICS 20,000+ Scene PB spatial-varying f”
Inverse Rendering for Complex Indoor Scenes: Shape, Spatially-Varying Lighting and SVBRDF From a Single Image (2019, methodology): “Previous work recognizes materials in indoor scenes [9] and decomposes indoor images into reflectance and shading layers [8, 31].”
Neural Inverse Rendering of an Indoor Scene From a Single Image (2019, methodology): “We compare with SIRFS on the test data from the IIW dataset [2].”
Shading Annotations in the Wild (2017, methodology): “Reflectance values tend to be sparse in indoor scenes, due to the overwhelming presence of human-made objects, which is often used as a prior in the intrinsic image literature [14, 5, 37].”
Learning Sparse High Dimensional Filters: Image Filtering, Dense CRFs and Bilateral Neural Networks (2015, background): “The dense CRF models have found wide-spread use in various computer vision applications [43, 10, 49, 47, 46, 11].”
OpenRooms: An End-to-End Open Framework for Photorealistic Indoor Scene Datasets (2020, background): “Ours + IIW 16.4 Li18[28] CGI + IIW 17.5 Sen.19[38] CGP + IIW 16.7 Li20[29] CGM + IIW 15.9”
SceneNN: A Scene Meshes Dataset with aNNotations (2016, methodology): “We apply the Retinex method for mesh vertices which is one of the simplest but effective decomposition algorithm [3].”