Layered object detection for multi-class segmentation

By Yi Yang, Sam Hallman, Deva Ramanan, Charless C. Fowlkes. 2010

114 citations recorded by the source corpus.

We formulate a layered model for object detection and multi-class segmentation. Our system uses the output of a bank of object detectors in order to define shape priors for support masks and then estimates appearance, depth ordering and labeling of pixels in the image. We train our system on the PASCAL segmentation challenge dataset and show good test results with state of the art performance in several categories including segmenting humans.

Subfields: Aerospace Engineering, 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 Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Convolutional feature masking for joint object and stuff segmentation (2014, background): “Semantic segmentation [14, 19, 24, 2] aims to label each image pixel to a semantic category.”
Learning to localize detected objects (2012, methodology): “Ramanan and colleagues [18, 25] incorporate position pri-”
An Exemplar-Based CRF for Multi-instance Object Segmentation (2014, background): “[27] they do not assume a depth ordering but can still handle partial occlusions.”
Nonparametric Scene Parsing via Label Transfer (2011, background): “Moreover, the authors in [47] showed that with a database of 80 million images, even simple SSD match can give semantically meaningful parsing for 32 32 images.”
Describing the scene as a whole: Joint object detection, scene classification and semantic segmentation (2012, background): “However, they either tackle only a subset of tasks [24, 6, 25, 15, 27], treat different components as black boxes [9], and/or rely on complex inference algorithms which are difficult to extend to incorporate other types of potentials/tasks [15].”
Diagnosing Error in Object Detectors (2012, background): “, [8, 25]) tend to combine simple shape and/or color priors with more generic segmentation techniques, such as graph cuts.”
Object detection by labeling superpixels (2015, background): “[22, 58, 59] infer whether an off-the-shelf detection is right or wrong by jointly optimize the detection and segmentation.”
Who Blocks Who: Simultaneous clothing segmentation for grouping images (2011, methodology): “Relationship The optimal blocking relationship˜b can be obtained by maximizing the individual local blocking distribution, i.e. ˜ b = (i,j)∈E arg max bij P (b ij |x i , x j ) (11) Yang et al. [22] search the optimal ordering from all possible permutations.”
Multi-scale context for scene labeling via flexible segmentation graph (2016, background): “Scene labeling is related to many applications, such as object detection [1,2] and recognition [3], visual attention [4], scene understanding [5,6], medical image segmentation [7], image alignment [8] and matching [9–11].”
Constrained parametric min-cuts for automatic object segmentation (2010, background): “Semantic segmentation algorithms [10], [11], [12], [13] incorporate the outputs of these object detector s, and may need to mediate the transition between the rectangular regions produced by the detector and the desired free-form”
Object segmentation by alignment of poselet activations to image contours (2011, background): “Most related to our work is the approach in [19].”
Semantic segmentation using regions and parts (2012, methodology): “, [31, 20, 1, 27, 15, 10, 37, 5, 12, 21]; the second one is addressed naturally in a multiscale scanning window detec-”
Harmony Potentials (2011, background): “For instance, Yang et al (2010) merge several object detections by layering the scene, and infers which object is in front of the other.”
Semantic object segmentation via detection in weakly labeled video (2015, methodology): “Instead, the image-based object detectors, which have demonstrated great successes in segmenting semantic objects in images [32, 17, 14, 26, 31], are employed into our segmentationby-detection framework.”
SLIC Superpixels Compared to State-of-the-Art Superpixel Methods (2012, unknown): “They have become key building blocks of many computer vision algorithms, such as top scoring multiclass object segmentation entries to the PASCAL VOC Challenge [9], [29], [11], depth estimation [30], segmentation [16], body model estimation [22], and object localization [9].”
Semantic Amodal Segmentation (2015, background): “Finally there has been some algorithmic work on amodal completion [14, 15, 38, 19] and depth ordering [41, 45].”
CPMC: Automatic Object Segmentation Using Constrained Parametric Min-Cuts (2012, background): “Semantic segmentation algorithms [11], [12], [13], [14] incorporate the outputs of these object detectors, and may need to mediate the transition between the rectangular regions produced by the detector and the desired free-form regions that align with object boundaries.”