Single-Class Instance Segmentation for Vectorization of Line Drawings
Bibliographic record
Abstract
Images can be represented and stored either in raster or in vector formats. Raster images are the most ubiquitous and are defined as matrices of pixel intensities/colours, while vector images consist of a finite set of geometric primitives, such as lines, curves, and polygons. Since geometric shapes are expressed via mathematical equations and defined by a limited number of control points, they can be manipulated in a much easier way than by directly working with pixels; hence, the vector format is much preferred to raster for image editing and understanding purposes. The conversion of a raster image into its vector correspondent is a non-trivial process, called image vectorization. Creating vector images from a given raster image can be time-consuming and requires the expertise of a skilled graphic user. This thesis explores the effectiveness of a Deep Learning based framework to vectorize raster images comprising line drawings with minimal user interventions. To improve the visual representation of the image, each stroke in the line drawing is represented with a different label and vectorized. In this document, we present an in-depth study of image vectorization, the objective of our research, challenges, potential resolutions, and compare the outcomes of our approach on six datasets consisting of different types of hand drawings. More specifically, this thesis begins by comparing raster images with vector images, the importance of image vectorization, and our objective to convert raster images to vector-based representations by accurately separating each stroke from the line drawings. In further chapters of this thesis, a Deep Learning based segmentation methodology is introduced to perform Single-Class Instance Segmentation of hand drawings to process the input raster image by labeling each pixel as belonging to a particular stroke instance. This segmentation approach is able to leverage the spatial relationships between each stroke instance. A novel loss function specifically designed to optimize our highly imbalanced datasets by scaling the margins and adding a regularization term to improve its feature selection technique. The weighted combination of our proposed margin regularized loss function is combined with the Dice loss to reduce the spatial overlap and improve the predictions over infrequent labels. Finally, the effectiveness of our segmentation technique of line drawing vectorization is compared experimentally with the state-of-the-art and our reference method. Our method can successfully handle a wide variety of human drawing styles. The results are comparable in terms of accuracy and way ahead in terms of speed and complexity, with other methods.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.008 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".