Multiscale analysis of depth images from natural scenes: Scaling in the depth of the woods

Chene, Yann; Belin, Étienne; Rousseau, D.; Chapeau‐Blondeau, François

doi:10.1016/j.chaos.2013.07.007

Cited by 11 publications

(14 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This makes a strong prediction: if observers internalise these statistics, their prior for small slants should be stronger in a context where surfaces vary along the 0°–180° tilt axis than a context in which surfaces vary along the 90°–270° tilt axis. Prior studies53336495051 demonstrating bias toward frontoparallel have varied surfaces either along the meridional or the azimuthal axis, but not both, and since methods vary widely between studies it is difficult to compare them directly. This prediction thus remains to be tested.…”

Section: Discussionmentioning

confidence: 99%

“…For example, previous studies of 3D scene statistics have revealed that surfaces further from the observer tend to be darker3435, which might explain analogous perceptual biases measured in the laboratory363738. Other statistical studies have found that discontinuities are larger when the occluding contour is convex, and this relationship is mirrored by biases in human estimatation of depth39.…”

mentioning

confidence: 84%

“…Previous work suggests that human observers tend to underestimate surface slant: surfaces are seen as closer to fronto-parallel than they really are53336495051. This may be due in part to cue conflicts from fronto-parallel display screens27, but it may also be due to an adaptive bias learnt from the environment, if larger slants are encountered less frequently.…”

Section: The Syns Datasetmentioning

confidence: 99%

See 2 more Smart Citations

The Southampton-York Natural Scenes (SYNS) dataset: Statistics of surface attitude

Adams

Elder

Graf

et al. 2016

Sci Rep

117

View full text Add to dashboard Cite

Recovering 3D scenes from 2D images is an under-constrained task; optimal estimation depends upon knowledge of the underlying scene statistics. Here we introduce the Southampton-York Natural Scenes dataset (SYNS: https://syns.soton.ac.uk), which provides comprehensive scene statistics useful for understanding biological vision and for improving machine vision systems. In order to capture the diversity of environments that humans encounter, scenes were surveyed at random locations within 25 indoor and outdoor categories. Each survey includes (i) spherical LiDAR range data (ii) high-dynamic range spherical imagery and (iii) a panorama of stereo image pairs. We envisage many uses for the dataset and present one example: an analysis of surface attitude statistics, conditioned on scene category and viewing elevation. Surface normals were estimated using a novel adaptive scale selection algorithm. Across categories, surface attitude below the horizon is dominated by the ground plane (0° tilt). Near the horizon, probability density is elevated at 90°/270° tilt due to vertical surfaces (trees, walls). Above the horizon, probability density is elevated near 0° slant due to overhead structure such as ceilings and leaf canopies. These structural regularities represent potentially useful prior assumptions for human and machine observers, and may predict human biases in perceived surface attitude.

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 84%

Section: The Syns Datasetmentioning

confidence: 99%

See 1 more Smart Citation

The Southampton-York Natural Scenes (SYNS) dataset: Statistics of surface attitude

Adams

Elder

Graf

et al. 2016

Sci Rep

117

View full text Add to dashboard Cite

show abstract

“…The depth map images from outdoor scenes of woods and plants as in [16,17] with the DSP( f ) of Eq. (1) were shown in [6] to also reveal self-similarity and fractal properties.…”

Section: Multiscale Frequency Analysismentioning

confidence: 99%

“…The colorimetric organization of natural landscapes with plants also demonstrates self-similarity and fractal properties [2,3,4] with other metrics based on the 3D RGB histogram. More recently, multiscale analysis has been undertaken for plant images obtained from another imaging technique delivering depth images of a physical scene [6]. The depth map images from outdoor scenes of woods and plants as in [16,17] with the DSP( f ) of Eq.…”

Section: Multiscale Frequency Analysismentioning

confidence: 99%

Counting leaves without ``finger-counting'' by supervised multiscale frequency analysis of depth images from top view

Rousseau

Zedde²

2015

Procedings of the Proceedings of the Computer Vision Problems in Plant Phenotyping Workshop 2015

Self Cite

View full text Add to dashboard Cite

Depth imaging is applied to characterize the shoot of seedlings from top-view. We demonstrate how quantitative informations of biological interest, such as leaves counting can be extracted from such images without performing 3D reconstruction of the shoot. This is obtained from 2D Fourier multiscale analysis without any requirement to segment nor detect leaves one by one numerically. We discuss the robustness and limitations of this approach and present possible extension with 3D Fourier analysis applied to estimate the plant plastochrone or 3D + T Fourier analysis in the estimation of circadian rythms.

show abstract

Imaging Methods for Phenotyping of Plant Traits

Rousseau

Dee²,

Pridmore

2015

Phenomics in Crop Plants: Trends, Options and Limitations

View full text Add to dashboard Cite

Multiscale analysis of depth images from natural scenes: Scaling in the depth of the woods

Cited by 11 publications

References 65 publications

The Southampton-York Natural Scenes (SYNS) dataset: Statistics of surface attitude

The Southampton-York Natural Scenes (SYNS) dataset: Statistics of surface attitude

Counting leaves without ``finger-counting'' by supervised multiscale frequency analysis of depth images from top view

Imaging Methods for Phenotyping of Plant Traits

Contact Info

Product

Resources

About