Himanshu Kumar

Mukund Agarwalla, Himanshu Kumar, Raj Dandekar et al.

Vision-Language Models (VLMs), such as GPT-4V and Llama 3.2 vision, have garnered significant research attention for their ability to leverage Large Language Models (LLMs) in multimodal tasks. However, their potential is constrained by inherent challenges, including proprietary restrictions, substantial computational demands, and limited accessibility. Smaller models, such as GIT and BLIP, exhibit marked limitations, often failing to generate coherent and consistent text beyond a few tokens, even with extensive training. This underscores a pivotal inquiry: how small can a VLM be and still produce fluent and consistent text? Drawing inspiration from the exceptional learning process of 3-4 year old children, who rely heavily on visual cues for understanding and communication, we introduce two novel datasets: ShortDesc (featuring concise image descriptions) and LongDesc (containing more detailed image descriptions). These datasets consist of image-text pairs where the text is restricted to the simple vocabulary and syntax typically used by young children, generated with a scaled-down model, GPT-4o. Using these datasets, we demonstrate that it is possible to train VLMs that are significantly smaller, up to 10 times smaller than state of the art(SOTA) small VLMs while maintaining architectural simplicity. To evaluate the outputs, we leverage GPT-4o to grade the text, as if stories written by students, on creativity, meaningfulness, and consistency, assigning scores out of 10. This method addresses limitations of standard benchmarks by accommodating unstructured outputs and providing a multidimensional evaluation of the model capabilities. Our findings contribute to the development of lightweight, accessible multimodal models for resource constrained environments.

Preeti Meena, Himanshu Kumar, Sandeep Yadav

Image summary, an abridged version of the original visual content, can be used to represent the scene. Thus, tasks such as scene classification, identification, indexing, etc., can be performed efficiently using the unique summary. Saliency is the most commonly used technique for generating the relevant image summary. However, the definition of saliency is subjective in nature and depends upon the application. Existing saliency detection methods using RGB-D data mainly focus on color, texture, and depth features. Consequently, the generated summary contains either foreground objects or non-stationary objects. However, applications such as scene identification require stationary characteristics of the scene, unlike state-of-the-art methods. This paper proposes a novel volumetric saliency-guided framework for indoor scene classification. The results highlight the efficacy of the proposed method.

Sourya Dipta Das, Nisarg A. Shah, Saikat Dutta et al.

Custom and natural lighting conditions can be emulated in images of the scene during post-editing. Extraordinary capabilities of the deep learning framework can be utilized for such purpose. Deep image relighting allows automatic photo enhancement by illumination-specific retouching. Most of the state-of-the-art methods for relighting are run-time intensive and memory inefficient. In this paper, we propose an efficient, real-time framework Deep Stacked Relighting Network (DSRN) for image relighting by utilizing the aggregated features from input image at different scales. Our model is very lightweight with total size of about 42 MB and has an average inference time of about 0.0116s for image of resolution $1024 \times 1024$ which is faster as compared to other multi-scale models. Our solution is quite robust for translating image color temperature from input image to target image and also performs moderately for light gradient generation with respect to the target image. Additionally, we show that if images illuminated from opposite directions are used as input, the qualitative results improve over using a single input image.