arXiv subjects

Ripon Saha

Publications and source records attributed to Ripon Saha.

6 recordsLinked to original sources

arXiv2026

Fine-Tuning a Large Vision-Language Model for Artwork's Scoring and Critique

Assessing artistic creativity is foundational to creativity research and arts education, yet manual scoring (e.g., Torrance Tests of Creative Thinking) is labor-intensive at scale. Prior machine-learning approaches show promise for visual creativity scoring, but many rely mainly on image features and provide limited or no explanatory feedback. We propose a framework for automated creativity assessment of human paintings by fine-tuning the vision-language model Qwen2-VL-7B with multi-task learning. Our dataset contains 1000 human-created paintings scored on a 1-100 scale and paired with a short human-written description (content or artist explanation). Two expert raters evaluated each work using a five-dimension rubric (originality, color, texture, composition, content) and provided written critiques; we use an 80/20 train-test split. We add a lightweight regression head on the visual encoder output so the model can predict a numerical score and generate rubric-aligned feedback in a single forward pass. By embedding the structured rubric and the artwork description in the system prompt, we constrain the generated text to match the quantitative prediction. Experiments show strong accuracy, achieving Pearson r > 0.97 and MAE about 3.95 on the 100-point scale. Qualitative evaluation indicates the generated feedback is semantically close to expert critiques (average SBERT cosine similarity = 0.798). The proposed approach bridges computer vision and art assessment and offers a scalable tool for creativity research and classroom feedback.

cs.CV↗

arXiv2024

Using a CNN Model to Assess Paintings' Creativity

Assessing artistic creativity has long challenged researchers, with traditional methods proving time-consuming. Recent studies have applied machine learning to evaluate creativity in drawings, but not paintings. Our research addresses this gap by developing a CNN model to automatically assess the creativity of human paintings. Using a dataset of six hundred paintings by professionals and children, our model achieved 90% accuracy and faster evaluation times than human raters. This approach demonstrates the potential of machine learning in advancing artistic creativity assessment, offering a more efficient alternative to traditional methods.

cs.CV↗

arXiv2024

The Massive and Distant Clusters of WISE Survey 2: Equatorial First Data Release

The Massive and Distant Clusters of WISE Survey 2 (MaDCoWS2) is a new survey designed as the successor of the original MaDCoWS survey. MaDCoWS2 improves upon its predecessor by using deeper optical and infrared data and a more powerful detection algorithm (PZWav). As input to the search, we use grz photometry from DECaLS in combination with W1 and W2 photometry from the CatWISE2020 catalog to derive the photometric redshifts with full redshift probability distribution functions for WISE-selected galaxies. Cluster candidates are then detected using the PZWav algorithm to find three-dimensional galaxy overdensities from the sky positions and photometric redshifts. This paper provides the first MaDCoWS2 data release, covering 1461 (1838 without masking) deg^2 centered on the Hyper-SuprimeCam Subaru Strategic Program equatorial fields. Within this region, we derive a catalog of 22,970 galaxy cluster candidates detected at S/N>5. These clusters span the redshift range 0.1 1.5. We compare MaDCoWS2 to six existing catalogs in the area. We rediscover 60%-92% of the clusters in these surveys at S/N>5. The medians of the absolute redshift offset are <0.02 relative to these surveys, while the standard deviations are less than 0.06. The median offsets between the detection position from MaDCoWS2 and other surveys are less than 0.25 Mpc. We quantify the relation between S/N and gas mass, total mass, luminosity, and richness from other surveys using a redshift-dependent power law relation. We find that the S/N-richness relation exhibits the lowest scatter.

astro-ph.CO↗

arXiv2023

Unsupervised Region-Growing Network for Object Segmentation in Atmospheric Turbulence

Moving object segmentation in the presence of atmospheric turbulence is highly challenging due to turbulence-induced irregular and time-varying distortions. In this paper, we present an unsupervised approach for segmenting moving objects in videos downgraded by atmospheric turbulence. Our key approach is a detect-then-grow scheme: we first identify a small set of moving object pixels with high confidence, then gradually grow a foreground mask from those seeds to segment all moving objects. This method leverages rigid geometric consistency among video frames to disentangle different types of motions, and then uses the Sampson distance to initialize the seedling pixels. After growing per-frame foreground masks, we use spatial grouping loss and temporal consistency loss to further refine the masks in order to ensure their spatio-temporal consistency. Our method is unsupervised and does not require training on labeled data. For validation, we collect and release the first real-captured long-range turbulent video dataset with ground truth masks for moving objects. Results show that our method achieves good accuracy in segmenting moving objects and is robust for long-range videos with various turbulence strengths.

cs.CV↗

arXiv2021

The Massive and Distant Clusters of WISE Survey XI: Stellar Mass Fractions and Luminosity Functions of MaDCoWS Clusters at $z \sim 1$

We present stellar mass fractions and composite luminosity functions (LFs) for a sample of \Ncl\ clusters from the Massive and Distant Clusters of WISE Survey (MaDCoWS) at a redshift range of $0.951 \leq z \leq 1.43$. Using SED fitting of optical and deep mid-infrared photometry, we establish the membership of objects along the lines-of-sight to these clusters and calculate the stellar masses of member galaxies. We find stellar mass fractions for these clusters largely consistent with previous works, including appearing to display a negative correlation with total cluster mass. We measure a composite $3.6~\mathrm{μm}$ LF down to $m^*+2.5$ for all 12 clusters. Fitting a Schechter function to the LF, we find a characteristic $3.6~\mathrm{μm}$ magnitude of $m^*=19.83\pm0.12$ and faint-end slope of $α=-0.81\pm0.10$ for the full sample at a mean redshift of $\bar{z} = 1.18$. We also divide the clusters into high- and low-redshift bins at $\bar{z}=1.29$ and $\bar{z}=1.06$ respectively and measure a composite LF for each bin. We see a small, but statistically significant evolution in $m^*$ and $α$ -- consistent with passive evolution -- when we study the joint fit to the two parameters, which is probing the evolution of faint cluster galaxies at $z\sim1$. This highlights the importance of deep IR data in studying the evolution of cluster galaxy populations at high-redshift.

astro-ph.GA↗

arXiv2018

Entropy Guided Spectrum Based Bug Localization Using Statistical Language Model

Locating bugs is challenging but one of the most important activities in software development and maintenance phase because there are no certain rules to identify all types of bugs. Existing automatic bug localization tools use various heuristics based on test coverage, pre-determined buggy patterns, or textual similarity with bug report, to rank suspicious program elements. However, since these techniques rely on information from single source, they often suffer when the respective source information is inadequate. For instance, the popular spectrum based bug localization may not work well under poorly written test suite. In this paper, we propose a new approach, EnSpec, that guides spectrum based bug localization using code entropy, a metric that basically represents naturalness of code derived from a statistical language model. Our intuition is that since buggy code are high entropic, spectrum based bug localization with code entropy would be more robust in discriminating buggy lines vs. non-buggy lines. We realize our idea in a prototype, and performed an extensive evaluation on two popular publicly available benchmarks. Our results demonstrate that EnSpec outperforms a state-of-the-art spectrum based bug localization technique.

cs.SE↗