Considering Human Perception and Memory in Interactive Multimedia Retrieval Evaluations
نویسندگان
چکیده
Experimental evaluations dealing with visual known-item search tasks, where real users look for previously observed and memorized scenes in a given video collection, represent challenging methodological problem. Playing searched “known” scene to prior the task start may not be sufficient terms of memorization re-identification (i.e., need necessarily successfully “implanted”). On other hand, enabling observe known played loop lead unrealistic situations can exploit very specific details that would remain their memory common case. To address these issues, we present proof-of-concept implementation new presentation methodology relies on recently introduced deep saliency estimation method limit amount revealed contents. A filtering process predicts subsequently removes information which an unconstrained setting likely leave lasting impression human observer. The proposed is compliant realistic assumption perceive memorize only limited information, at same time allows play verification purposes. also serves as clue equalizer, limiting rich set exploitable content features thus unifies perceived by different users. performed evaluation demonstrates feasibility such showing retrieval still possible based query videos processed method. We postulate incomplete tasks constitute necessary next step challenge assess interactive multimedia systems participating campaigns.
منابع مشابه
Interactive Indexing and Retrieval of Multimedia Content
The indexing and retrieval of multimedia items is difficult due to the semantic gap between the user’s perception of the data and the descriptions we can derive automatically from the data using computer vision, speech recognition, and natural language processing. In this contribution we consider the nature of the semantic gap in more detail and show examples of methods that help in limiting th...
متن کاملAdaptive Pattern Discovery for Interactive Multimedia Retrieval
Relevance feedback has been an indispensable component for multimedia retrieval systems. In this paper, we present an adaptive pattern discovery method, which addresses relevance feedback by interactively discovering meaningful patterns of relevant objects. To facilitate pattern discovery, we first present a dynamic feature extraction method, which aims to alleviate the curse of dimensionality ...
متن کاملLarge Scale Evaluations of Multimedia Information Retrieval: The TRECVid Experience
Information Retrieval is a supporting technique which underpins a broad range of content-based applications including retrieval, filtering, summarisation, browsing, classification, clustering, automatic linking, and others. Multimedia information retrieval (MMIR) represents those applications when applied to multimedia information such as image, video, music, etc. In this presentation and exten...
متن کاملImagery-based Image Retrieval Considering Individual’s Subjective Perception
In this paper, we proposed an approach of imagery-based image retrieval considering individual’s subjective perception. By using eigen SGLD (Space Gray Level Dependence) matrices, the statistical and textural features of images were extracted, the associations between these features and semantic features based on human visual perception were analyzed, and the mapping models were designed to bri...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Lecture Notes in Computer Science
سال: 2021
ISSN: ['1611-3349', '0302-9743']
DOI: https://doi.org/10.1007/978-3-030-67832-6_49