Object segmentation by alignment of poselet activations to image contours

By Thomas Brox, Lubomir D. Bourdev, Subhransu Maji, Jitendra Malik. 2011

116 citations recorded by the source corpus.

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

Microsoft COCO: Common Objects in Context (2014, methodology): “Generating segmentations from detections We now describe a simple method for generating object bounding boxes and segmentation masks, following prior work that produces segmentations from object detections [47], [48], [49], [50].”
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.”
Lifting Object Detection Datasets into 3D (2015, methodology): “In particular, if the estimated shape was the correct one, all the keypoints, even the ones which are not visible, should reproject inside the silhouette.”
Iterative Instance Segmentation (2015, background): “This can be achieved simply by aligning the predicted segmentation to image contours [5] or projecting to superpixels [16].”
SCALPEL: Segmentation Cascades with Localized Priors and Efficient Learning (2013, methodology): “Using the annotations provided by [4], we first cluster all objects in the PASCAL VOC2011 training set into five aspect ratio clusters using K-means.”
Adaptive region pooling for object detection (2015, methodology): “First, we evaluate on the keypoint annotation dataset of [6].”
A High Performance CRF Model for Clothes Parsing (2014, background): “Part-based detectors have also been aligned to image contours to aid in object segmentation [24].”
Towards Unified Object Detection and Semantic Segmentation (2014, unknown): “employed Poselet detectors to predict masks for object parts [5].”
Is object localization for free? - Weakly-supervised learning with convolutional neural networks (2015, unknown): “Fully supervised methods [18] require careful annotation of object location in the form of bounding boxes [18], segmentation [58] or even location of object parts [5], which is costly and can introduce biases.”
Analyzing Semantic Segmentation Using Hybrid Human-Machine CRFs (2013, unknown): “depth estimation, object segmentation) [17, 16, 22, 5, 13].”
Semantic segmentation using regions and parts (2012, methodology): “[9] align a predicted object mask to the image contours, apply variational smoothing and assign figure/ground superpixels using self-similarity.”
Symbiotic Segmentation and Part Localization for Fine-Grained Categorization (2013, background): “In the poselet line of research [6] the detectors are for parts, rather than for entire categories, but again the poseletdetectors can predict foreground masks for object category detection and segmentation [9, 20].”
Shape Sharing for Object Segmentation (2012, methodology): “In segmentation, shape is typically used as a category-specific cue, whereby known object models are integrated with bottom-up grouping cues [6, 7, 11, 12].”
An Exemplar-Based CRF for Multi-instance Object Segmentation (2014, methodology): “In [5, 12], object detector outputs are integrated into a pixel-level CRF model to impose a soft top-down constraint.”
Using k-Poselets for Detecting People and Localizing Their Keypoints (2014, background): “A poselet’s training set is associated with a consistent keypoint configuration (unlike in DPM training), so by design when there is a strong activation of a poselet in a test image, one can make considerably tighter predictions that are useful for analyzing attributes [4], segmenting objects [6], r”
Learning to localize detected objects (2012, methodology): “[16] use only color models; others use template weights to identify object pixels or edge features [4, 11].”