Abhinav Ravi

Suranjan Goswami, Abhinav Ravi, Raja Kolla et al.

Despite significant advances in document understanding, determining the correct orientation of scanned or photographed documents remains a critical pre-processing step in the real world settings. Accurate rotation correction is essential for enhancing the performance of downstream tasks such as Optical Character Recognition (OCR) where misalignment commonly arises due to user errors, particularly incorrect base orientations of the camera during capture. In this study, we first introduce OCR-Rotation-Bench (ORB), a new benchmark for evaluating OCR robustness to image rotations, comprising (i) ORB-En, built from rotation-transformed structured and free-form English OCR datasets, and (ii) ORB-Indic, a novel multilingual set spanning 11 Indic mid to low-resource languages. We also present a fast, robust and lightweight rotation classification pipeline built on the vision encoder of Phi-3.5-Vision model with dynamic image cropping, fine-tuned specifically for 4-class rotation task in a standalone fashion. Our method achieves near-perfect 96% and 92% accuracy on identifying the rotations respectively on both the datasets. Beyond classification, we demonstrate the critical role of our module in boosting OCR performance: closed-source (up to 14%) and open-weights models (up to 4x) in the simulated real-world setting.

Shaharukh Khan, Ali Faraz, Abhinav Ravi et al.

Multimodal research has predominantly focused on single-image reasoning, with limited exploration of multi-image scenarios. Recent models have sought to enhance multi-image understanding through large-scale pretraining on interleaved image-text datasets. However, most Vision-Language Models (VLMs) are trained primarily on English datasets, leading to inadequate representation of Indian languages. To address this gap, we introduce the Chitrakshara dataset series, covering 11 Indian languages sourced from Common Crawl. It comprises (1) Chitrakshara-IL, a large-scale interleaved pretraining dataset with 193M images, 30B text tokens, and 50M multilingual documents, and (2) Chitrakshara-Cap, which includes 44M image-text pairs with 733M tokens. This paper details the data collection pipeline, including curation, filtering, and processing methodologies. Additionally, we present a comprehensive quality and diversity analysis to assess the dataset's representativeness across Indic languages and its potential for developing more culturally inclusive VLMs.

Ali Faraz, Akash, Shaharukh Khan et al.

Vision-language models (VLMs) have demonstrated impressive generalization across multimodal tasks, yet most evaluation benchmarks remain Western-centric, leaving open questions about their performance in culturally diverse and multilingual settings. To address this gap, we introduce IndicVisionBench, the first large-scale benchmark centered on the Indian subcontinent. Covering English and 10 Indian languages, our benchmark spans 3 multimodal tasks, including Optical Character Recognition (OCR), Multimodal Machine Translation (MMT), and Visual Question Answering (VQA), covering 6 kinds of question types. Our final benchmark consists of a total of ~5K images and 37K+ QA pairs across 13 culturally grounded topics. In addition, we release a paired parallel corpus of annotations across 10 Indic languages, creating a unique resource for analyzing cultural and linguistic biases in VLMs. We evaluate a broad spectrum of 8 models, from proprietary closed-source systems to open-weights medium and large-scale models. Our experiments reveal substantial performance gaps, underscoring the limitations of current VLMs in culturally diverse contexts. By centering cultural diversity and multilinguality, IndicVisionBench establishes a reproducible evaluation framework that paves the way for more inclusive multimodal research.

Shaharukh Khan, Ayush Tarun, Abhinav Ravi et al.

Recent multimodal foundation models are primarily trained on English or high resource European language data, which hinders their applicability to other medium and low-resource languages. To address this limitation, we introduce Chitrarth (Chitra: Image; Artha: Meaning), an inclusive Vision-Language Model (VLM), specifically targeting the rich linguistic diversity and visual reasoning across 10 prominent Indian languages. Our model effectively integrates a state-of-the-art (SOTA) multilingual Large Language Model (LLM) with a vision module, primarily trained on multilingual image-text data. Furthermore, we also introduce BharatBench, a comprehensive framework for evaluating VLMs across various Indian languages, ultimately contributing to more diverse and effective AI systems. Our model achieves SOTA results for benchmarks across low resource languages while retaining its efficiency in English. Through our research, we aim to set new benchmarks in multilingual-multimodal capabilities, offering substantial improvements over existing models and establishing a foundation to facilitate future advancements in this arena.

Shaharukh Khan, Ayush Tarun, Ali Faraz et al.

In this work, we provide the system description of our submission as part of the English to Lowres Multimodal Translation Task at the Workshop on Asian Translation (WAT2024). We introduce Chitranuvad, a multimodal model that effectively integrates Multilingual LLM and a vision module for Multimodal Translation. Our method uses a ViT image encoder to extract visual representations as visual token embeddings which are projected to the LLM space by an adapter layer and generates translation in an autoregressive fashion. We participated in all the three tracks (Image Captioning, Text only and Multimodal translation tasks) for Indic languages (ie. English translation to Hindi, Bengali and Malyalam) and achieved SOTA results for Hindi in all of them on the Challenge set while remaining competitive for the other languages in the shared task.

Aditya Kallappa, Palash Kamble, Abhinav Ravi et al.

India is a diverse society with unique challenges in developing AI systems, including linguistic diversity, oral traditions, data accessibility, and scalability. Existing foundation models are primarily trained on English, limiting their effectiveness for India's population. Indic languages comprise only 1 percent of Common Crawl corpora despite India representing 18 percent of the global population, leading to linguistic biases. Thousands of regional languages, dialects, and code mixing create additional representation challenges due to sparse training data. We introduce Krutrim LLM, a 2 trillion token multilingual model designed for India's linguistic landscape. It incorporates the largest known Indic dataset, mitigating data scarcity and ensuring balanced performance across dialects. Krutrim outperforms or matches state-of-the-art models on Indic benchmarks while maintaining competitive English performance. Despite being significantly smaller in training flops, Krutrim LLM matches or exceeds models like LLAMA-2 on 10 out of 16 tasks, with an average score of 0.57 versus 0.55. This evidences Krutrim's flexible multilingual fluency across diverse linguistic contexts. Krutrim is integrated with real-time search to improve factual accuracy in conversational AI applications. This enhances accessibility for over 1 billion users worldwide. Through intentional design choices addressing data imbalances, Krutrim LLM signifies meaningful progress in building ethical, globally representative AI models.

Ujjal Kr Dutta, Sandeep Repakula, Maulik Parmar et al.

In this paper, we address a crucial problem in fashion e-commerce (with respect to customer experience, as well as revenue): color variants identification, i.e., identifying fashion products that match exactly in their design (or style), but only to differ in their color. We propose a generic framework, that leverages deep visual Representation Learning at its heart, to address this problem for our fashion e-commerce platform. Our framework could be trained with supervisory signals in the form of triplets, that are obtained manually. However, it is infeasible to obtain manual annotations for the entire huge collection of data usually present in fashion e-commerce platforms, such as ours, while capturing all the difficult corner cases. But, to our rescue, interestingly we observed that this crucial problem in fashion e-commerce could also be solved by simple color jitter based image augmentation, that recently became widely popular in the contrastive Self-Supervised Learning (SSL) literature, that seeks to learn visual representations without using manual labels. This naturally led to a question in our mind: Could we leverage SSL in our use-case, and still obtain comparable performance to our supervised framework? The answer is, Yes! because, color variant fashion objects are nothing but manifestations of a style, in different colors, and a model trained to be invariant to the color (with, or without supervision), should be able to recognize this! This is what the paper further demonstrates, both qualitatively, and quantitatively, while evaluating a couple of state-of-the-art SSL techniques, and also proposing a novel method.

Ujjal Kr Dutta, Sandeep Repakula, Maulik Parmar et al.

In this paper, we utilize deep visual Representation Learning to address an important problem in fashion e-commerce: color variants identification, i.e., identifying fashion products that match exactly in their design (or style), but only to differ in their color. At first we attempt to tackle the problem by obtaining manual annotations (depicting whether two products are color variants), and train a supervised triplet loss based neural network model to learn representations of fashion products. However, for large scale real-world industrial datasets such as addressed in our paper, it is infeasible to obtain annotations for the entire dataset, while capturing all the difficult corner cases. Interestingly, we observed that color variants are essentially manifestations of color jitter based augmentations. Thus, we instead explore Self-Supervised Learning (SSL) to solve this problem. We observed that existing state-of-the-art SSL methods perform poor, for our problem. To address this, we propose a novel SSL based color variants model that simultaneously focuses on different parts of an apparel. Quantitative and qualitative evaluation shows that our method outperforms existing SSL methods, and at times, the supervised model.

Rajdeep Hazra Banerjee, Abhinav Ravi, Ujjal Kr Dutta

Popular fashion e-commerce platforms mostly provide details about low-level attributes of an apparel (eg, neck type, dress length, collar type) on their product detail pages. However, customers usually prefer to buy apparel based on their style information, or simply put, occasion (eg, party/ sports/ casual wear). Application of a supervised image-captioning model to generate style-based image captions is limited because obtaining ground-truth annotations in the form of style-based captions is difficult. This is because annotating style-based captions requires a certain amount of fashion domain expertise, and also adds to the costs and manual effort. On the contrary, low-level attribute based annotations are much more easily available. To address this issue, we propose a transfer-learning based image captioning model that is trained on a source dataset with sufficient attribute-based ground-truth captions, and used to predict style-based captions on a target dataset. The target dataset has only a limited amount of images with style-based ground-truth captions. The main motivation of our approach comes from the fact that most often there are correlations among the low-level attributes and the higher-level styles for an apparel. We leverage this fact and train our model in an encoder-decoder based framework using attention mechanism. In particular, the encoder of the model is first trained on the source dataset to obtain latent representations capturing the low-level attributes. The trained model is fine-tuned to generate style-based captions for the target dataset. To highlight the effectiveness of our method, we qualitatively and quantitatively demonstrate that the captions generated by our approach are close to the actual style information for the evaluated apparel. A Proof Of Concept for our model is under pilot at Myntra where it is exposed to some internal users for feedback.

Abhinav Ravi, Sandeep Repakula, Ujjal Kr Dutta et al.

Have you ever looked at an Instagram model, or a model in a fashion e-commerce web-page, and thought \textit{"Wish I could get a list of fashion items similar to the ones worn by the model!"}. This is what we address in this paper, where we propose a novel computer vision based technique called \textbf{ShopLook} to address the challenging problem of recommending similar fashion products. The proposed method has been evaluated at Myntra (www.myntra.com), a leading online fashion e-commerce platform. In particular, given a user query and the corresponding Product Display Page (PDP) against the query, the goal of our method is to recommend similar fashion products corresponding to the entire set of fashion articles worn by a model in the PDP full-shot image (the one showing the entire model from head to toe). The novelty and strength of our method lies in its capability to recommend similar articles for all the fashion items worn by the model, in addition to the primary article corresponding to the query. This is not only important to promote cross-sells for boosting revenue, but also for improving customer experience and engagement. In addition, our approach is also capable of recommending similar products for User Generated Content (UGC), eg., fashion article images uploaded by users. Formally, our proposed method consists of the following components (in the same order): i) Human keypoint detection, ii) Pose classification, iii) Article localisation and object detection, along with active learning feedback, and iv) Triplet network based image embedding model.

Abhinav Ravi, Arun Patro, Vikram Garg et al.

$ $"Fast Fashion" spearheads the biggest disruption in fashion that enabled to engineer resilient supply chains to quickly respond to changing fashion trends. The conventional design process in commercial manufacturing is often fed through "trends" or prevailing modes of dressing around the world that indicate sudden interest in a new form of expression, cyclic patterns, and popular modes of expression for a given time frame. In this work, we propose a fully automated system to explore, detect, and finally synthesize trends in fashion into design elements by designing representative prototypes of apparel given time series signals generated from social media feeds. Our system is envisioned to be the first step in design of Fast Fashion where the production cycle for clothes from design inception to manufacturing is meant to be rapid and responsive to current "trends". It also works to reduce wastage in fashion production by taking in customer feedback on sellability at the time of design generation. We also provide an interface wherein the designers can play with multiple trending styles in fashion and visualize designs as interpolations of elements of these styles. We aim to aid the creative process through generating interesting and inspiring combinations for a designer to mull by running them through her key customers.