Has My Algorithm Succeeded? An Evaluator for Human Pose Estimators

By Nataraj Jammalamadaka, Andrew Zisserman, Marcin Eichner, Vittorio Ferrari, C. V. Jawahar. 2012

How researchers see this paper

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

Citations in the corpus

Towards Open Set Deep Networks (2015, methodology): “Multi-class Meta-Recognition using Activation Vec- tors to estimate the probability of deep network failure 2.”
A Survey of Deep Learning-Based Object Detection (2019, unknown): “[56] A.”
Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Gradient Response Maps for Real-time Detection of Texture-less Objects (unknown): “From the related works which also take into account depth data there are mainly approaches related to pedestrian detection [26], [27], [28], [29].”
Global data association for multi-object tracking using network flows (2008, unknown): “The ETHMS dataset includes 4 video sequences of street scenes taken by a moving camera, with frame size of 640 × 480 and frame rate of 15fps.”
EuroCity Persons: A Novel Benchmark for Person Detection in Traffic Scenes (2018, unknown): “Daimler-DB [11], INRIA [19], ETH [20], and TUDBrussels [21]).”
Approximate Log-Hilbert-Schmidt Distances between Covariance Operators for Image Classification (2016, unknown): “From top to bottom: Fish Recognition [3], Virus Classiﬁcation [16], KTH-TIPS2b Material [5], Texture [4, 8], ETHZ [10], UIUC [20], Tiny-Graz03 [33], and ETH80 [18]. note the color values, and G o,sx,y denotes the Gabor ﬁlter at orientation o and scale s .”
Coupled Object Detection and Tracking from Static Cameras and Moving Vehicles (2008, unknown): “Other possible extensions include the integration of additional cues such as stereo depth [11] or the combination with adaptive background modeling for static cameras.”
A mobile vision system for robust multi-person tracking (2008, unknown): “Secondly, we remove the only explicit cycle between π and d i compared to [9], which simplifies the belief propagation algorithm used for inference and increases its stability.”
Learning scene-specific pedestrian detectors without real data (2015, unknown): “This stands in contrast to the large body of work focused primarily on pedestrian detection from a mobile platform [12, 13, 41, 11].”
Estimating contact dynamics (2009, unknown): “The motion of an object is determined in part by its contact with the environment, and conversely, motion is a rich source of information about contact, much like the locations of people are informative about the ground plane [8, 9].”
Object Detection and Tracking for Autonomous Navigation in Dynamic Environments (2010, unknown): “For details, see Ess et al. (2007).”
Bregman Divergences for Infinite Dimensional Covariance Matrices (2014, unknown): “Interestingly, it even outperforms the Table 5: Recognition accuracies for the ETHZ dataset [6].”
Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes (2011, unknown): “This technique combines image and depth data, and is mainly used for pedestrian detection [5] [6] [10] [27].”
Is Faster R-CNN Doing Well for Pedestrian Detection? (2016, unknown): “We comprehensively evaluate on 4 benchmarks: Caltech [14], INRIA [20], ETH [21] and KITTI [22].”
Fusion of Multispectral Data Through Illumination-aware Deep Neural Networks for Pedestrian Detection (2018, unknown): “Pedestrian detection is a popular research topic within the field of computer vision in the past decades [29, 5, 8, 11, 10, 4, 41].”
Recent Advances in Deep Learning for Object Detection (2019, unknown): “ETH [245] contains 1804 frames in three video clips and commonly it’s used as test set to evaluate performance of the models trained on the large scale datasets(CityPersons dataset etc.”
Person re-identification by symmetry-driven accumulation of local features (2010, unknown): “We tested our approach (in single-shot modality ) on three challenging public databases (VIPeR, iLIDS, and ETHZ), outperforming the highest performances on all but one dataset.”
Unsupervised Salience Learning for Person Re-identification (2013, unknown): “ETHZ Dataset [9].”
Human detection from images and videos: A survey (2016, unknown): “Grayscale (intensity) patterns were employed to represent the local appearance of parts of humans in [88, 32].”
Person Re-identification: Past, Present and Future (2016, unknown): “Several multi-shot re-ID datasets exist, e.g., ETH [122], 3DPES [123], PRID-2011 [86], iLIDS-VID [105], and MARS [21].”