Category-Specific Video Summarization

By Danila Potapov, Matthijs Douze, Zaïd Harchaoui, Cordelia Schmid. 2014

502 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Sociology and Political Science, Signal Processing

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

In Sociology and Political Science

In Signal Processing

Citations in the corpus

Learning Features by Watching Objects Move (2016, methodology): “We use a publicly available appearance-based shot detection method [40] (also unsupervised) to divide the video into shots and run uNLC separately on each shot.”
Unsupervised Learning from Narrated Instruction Videos (2015, background): “While summarization is a subjective task, we here aim to extract the key steps required to achieve a concrete task that consistently appear in the same sequence in the input set of videos.”
Unsupervised Semantic Parsing of Video Collections (2015, methodology): “Kernel Temporal Segmentation [48]: Kernel Temporal Segmentation (KTS) proposed by Potapov et al.”
Summarization of Egocentric Videos: A Comprehensive Survey (2017, background): “Furthermore, as many of the reviewed works point out [21], [23], [42], the ideal summary is context dependent.”
Enhancing Video Summarization via Vision-Language Embedding (2017, methodology): “Consequently, many methods for computing video summaries have been proposed by researchers [1, 5, 6, 8, 15, 20, 25, 30, 42, 48, 47].”
SCSampler: Sampling Salient Clips From Video for Efficient Action Recognition (2019, background): “For example, video summarization [18, 19, 29, 37, 56, 57, 58] and the automatic production of sport highlights [30, 31] entail creating a much shorter version of the original video by concatenating a small set of snippets corresponding to the most informative or exciting moments.”
Learning from Narrated Instruction Videos (2015, methodology): “Our work is also related to video summarization and in particular to the recent work on categoryspecific video summarization [16, 21].”
Making 360° Video Watchable in 2D: Learning Videography for Click Free Viewing (2017, background): “Video summarization Video summarization aims to generate a concise representation for a video by removing temporal redundancy while preserving the important events [8, 14–16, 24, 27, 32, 33, 38, 42].”
Video Summarization with Long Short-Term Memory (2016, methodology): “Several recent methods also exploit auxiliary information such as web images [10,11,33,35] or video categories [31] to facilitate the summarization process.”
A Survey of Deep Learning-Based Object Detection (2019, background): “Typical highlight detectors [256]–[261] are domain-specific for they are tailored to a category of videos.”
Temporal Localization of Fine-Grained Actions in Videos by Domain Transfer from Web Images (2015, methodology): “It was used by [32, 24] and achieved state-of-the-art performance in event recounting and video summarization tasks.”
Viewpoint-Aware Video Summarization (2018, background): “However, owing to the subjective property of this task, a supervised approach [21, 38, 32, 23, 12, 31, 13, 19, 9, 42], that trains a summarization model which takes human-created summaries as the supervision, became standard because of its better performance.”
Unsupervised Video Summarization with Adversarial LSTM Networks (2017, methodology): “To generate key shots for datasets which only provide key frame scores, the videos are initially temporally segmented into disjoint intervals using KTS [26].”
Revealing Event Saliency in Unconstrained Video Collection (2017, unknown): “Finally, to evaluate the proposed approach, we conduct experiments on three benchmarks that have shot-level annotations , including a collection in [8] and two collections recently released by the NIST TRECVID [9].”
DevNet: A Deep Event Network for multimedia event detection and evidence recounting (2015, background): “In [34, 25, 45, 30], they apply object and action detectors or low-level visual features to localize temporal key evidences.”
CLIP-It! Language-Guided Video Summarization (2021, methodology): “non-parametric unsupervised methods [14, 17, 24, 20, 21, 29] used various heuristics to represent the importance of frames.”