On the Temporality for Sketch Representation Learning
By: Marcelo Isaias de Moraes Junior, Moacir Antonelli Ponti
Potential Business Impact:
Makes computers understand drawings better by looking at order.
Sketches are simple human hand-drawn abstractions of complex scenes and real-world objects. Although the field of sketch representation learning has advanced significantly, there is still a gap in understanding the true relevance of the temporal aspect to the quality of these representations. This work investigates whether it is indeed justifiable to treat sketches as sequences, as well as which internal orders play a more relevant role. The results indicate that, although the use of traditional positional encodings is valid for modeling sketches as sequences, absolute coordinates consistently outperform relative ones. Furthermore, non-autoregressive decoders outperform their autoregressive counterparts. Finally, the importance of temporality was shown to depend on both the order considered and the task evaluated.
Similar Papers
An empirical study of the effect of video encoders on Temporal Video Grounding
CV and Pattern Recognition
Helps computers find video clips from descriptions.
Temporal Chunking Enhances Recognition of Implicit Sequential Patterns
Machine Learning (CS)
Teaches computers to learn faster from past experiences.
Thinking with Drafts: Speculative Temporal Reasoning for Efficient Long Video Understanding
CV and Pattern Recognition
Makes AI watch long videos faster and smarter.