Spatio-temporal Object Detection Proposals

By Dan Oneaţă, Jérôme Revaud, Jakob Verbeek, Cordelia Schmid. 2014

226 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

Object-Proposal Evaluation Protocol is ‘Gameable’ (2015, background): “Then average recall is measured over all the ground truth instances: where I [ · ] is an indicator function for the logical preposition in the argument.”
Towards Weakly-Supervised Action Localization (2016, background): “In parallel, several works [8,25] have attempted to further improve the quality of tubes.”
Unsupervised Object Discovery and Tracking in Video Collections (2015, methodology): “Finally, spatio-temporal proposals of [16, 24] and action localization [39, 42] are relevant to our work as they also return spatio-temporal tubes as output.”
Encouraging LSTMs to Anticipate Actions Very Early (2017, background): “This is in contrast to existing methods that typically extract either global representations for the entire image [6, 41, 42, 7] or video sequence [35, 14], thus not focusing on the action itself, or localize the feature extraction process to the action itself via dense trajectories [39, 38, 9], opt”
Human Action Localization with Sparse Spatial Supervision (2017, background): “In parallel, several works [11], [31] have attempted to further improve the quality of tubes.”
LIME: Low-Light Image Enhancement via Illumination Map Estimation (2017, background): “[3], [4], [5] can avoid the above problem by somehow forcing the output image to fall in the range [0] [1].”
Track and Segment: An Iterative Unsupervised Approach for Video Object Proposals (2016, methodology): “Finally, unlike [15, 27, 41, 38], which learn to propose objects either in videos or static images, our approach does not require any labeled video/image to train and instead adapts to each unknown object on a per-video basis.”
Temporal Context Network for Activity Localization in Videos (2017, methodology): “For spatio-temporal localization of actions, multiple methods use spatio-temporal region proposals [9, 22, 7, 39].”
FusionSeg: Learning to Combine Motion and Appearance for Fully Automatic Segmentation of Generic Objects in Videos (2017, methodology): “Second we have the fully automatic methods that generate thousands of “object-like” space-time segments [21], [22], [23], [24], [25].”
3D Object Proposals for Accurate Object Class Detection (2015, background): “Existing approaches range from purely RGB [4, 9, 10, 5, 15, 16], RGB-D [17, 14, 18, 19], to video [20].”
Learning to segment moving objects in videos (2014, background): “Work of [29] produces multiple video segments by canceling image boundaries that do not exhibit high ﬂow strength.”
Animal Detection From Highly Cluttered Natural Scenes Using Spatiotemporal Object Region Proposals and Patch Verification (2016, unknown): “2) Quantitative Results: We first compare the performance of our animal object proposal method using IEC with the following state-of-the-art object proposal methods: Spatial-Temporal Object Detection Proposals (STODP) [44], Fully Connected Object Proposals for Video Segmentation (FCOP) [43], and Lea”
T-CNN: Tubelets With Convolutional Neural Networks for Object Detection From Videos (2016, unknown): “and spatio-temporal object proposal algorithms [9].”
DAPs: Deep Action Proposals for Action Understanding (2016, methodology): “Spatio-Temporal Action Proposals: Recently, ideas from the area of object proposals have been extrapolated to action recognition in the video domain [6, 17, 11, 21, 24, 10, 39].”
Spatio-Temporal Moving Object Proposals (2014, background): “Approaches that condition on a static segmentation map to boost or suppress boundary fragments using optical ﬂow, such as [28,34,38], are upper-bounded by the performance of the static boundary detector: for high thresholds, many boundaries are missed with no hope to be recovered; for low…”
TORNADO: A Spatio-Temporal Convolutional Regression Network for Video Action Proposal (2017, background): “However, segmentation-and-merging [11, 22, 39, 26] and human detection [42] approaches rely on lowlevel cues which fall behind the state-of-the-art.”
What do 15,000 object categories tell us about classifying and localizing actions? (2015, unknown): “Though most of the emphasis in the action literature has been on classiﬁcation, progress has also been made in action localization [2, 15, 22, 29, 44].”
Monocular 3D Object Detection for Autonomous Driving (2016, background): “Most of the existing work on proposal generation uses RGB [45, 55, 9, 2, 11, 29], RGB-D [4, 21, 31, 25], or video [35].”
Recognizing Actions Through Action-Specific Person Detection (2015, background): “The use of generic object proposals [1], [5], [30], [40] for object detection has gained significant attention in recent years.”