ASYSST: A Framework for Synopsis Synthesis Empowering Visually Impaired
Citations
11 citations
Cites methods from "ASYSST: A Framework for Synopsis Sy..."
...In [14], [15], authors have used handcrafted features for identifying decor symbol, room information and generating region wise caption generation....
[...]
...1) Template based: Paragraph based descriptions are generated by using technique proposed in [14]....
[...]
6 citations
4 citations
References
21,126 citations
14,245 citations
13,993 citations
"ASYSST: A Framework for Synopsis Sy..." refers methods in this paper
...To delineate room boundaries, we detect doors using scale invariant features [9] and close the gaps in wall image corresponding to the door locations....
[...]
9,293 citations
"ASYSST: A Framework for Synopsis Sy..." refers methods in this paper
...Since ROUGE-1, ROUGE-2, and ROUGE-3 use uni-gram, bi-gram and trigram comparisons respectively, the decreasing nature of average precision is natural....
[...]
...We have compared the machine generated description of the floor plan with human written descriptions using 3metrics, ROUGE [12], BLEU [16] and METEOR [6]....
[...]
...As the value of n in n-gram comparison increasing, the ROUGE precision score decreases, which is also clear from Tab....
[...]
...Table 3 depicts the average recall, average precision and F score for ROUGE-1, ROUGE-2, ROUGE-3....
[...]
...We have compared the machine generated description of the floor plan with human written descriptions using 3metrics, ROUGE [12], BLEU [16] and METEOR [6]....
[...]
7,963 citations
"ASYSST: A Framework for Synopsis Sy..." refers background or methods in this paper
...Table 1 shows the quantitative comparison between ours, [4], [10] and our technique using LBP (local binary pattern) feature [15]....
[...]
...Also in [10], recognition is 0 for many decor items with low accuracy for others....
[...]