Maximizing cart value by increasing the number of items in electronic carts is one of the key strategies adopted by e-commerce platforms for optimal conversion of positive user intent during an online shopping session. Recommender systems play a key-role in suggesting personalized candidate items that can be added to cart by the user. However, it is important to serve a diverse set of personalized recommendations that 'complement' user's cart content to practically increase item count in cart and also contribute towards product discovery. Borrowed from Quantum Physics, Determinantal Point Processes (DPP) are used widely in recommender systems to diversify personalized product recommendations for improved user engagement. However, vertically scaling DPP for recommendation sets, personalized with vector similarity metric like cosine similarity, to serve large scale real-time concurrent user requests is non-trivial. We propose a vectorized reformulation of cosine similarity and conditional DPP implementation to best utilize the highly improved vector computation capabilities (SIMD) of modern processors. Experimental evidence on real-world traffic shows that the proposed method can handle upto 15x more concurrent traffic while improving latency. The proposed method also uses portable SIMD constructs from Python libraries which can be easily adopted in most available SIMD supported CPUs with minimal code changes.

Fashion Ecommerce is challenging because users have intents that could be very broad Eg: ''I want a dress to wear for my cousin's wedding''. Traditional single-shot retrieval mechanisms like search don't work very well for such intents. To solve this problem we developed ''Maya'', a conversational shopping assistant based on LLMs. Maya has been deployed in production and is available to select cohort of users at scale who come to our platform. It is arguably, one of the earliest production deployments of GenAI Fashion assistants at this scale. We describe the journey of building Maya and lessons post the production deployment.

Existing works on image deblurring operate either with an assumption of white Gaussian noise or ignore the noise term completely in formulating the blur generation process. However, real world images may additionally be corrupted with noise, compression and scaling artifacts, detail loss, etc. The state of the art image deblurring methods do not work well with real images due to this issue. In this work, we propose a blur generation formula that considers composite nature of real world artifacts, enabling deep learning methods to learn an end-to-end mapping between real world blurred and realistic sharp images. We propose an extremely light-weight multiscale-multiresidue architecture, designed for efficient execution on embedded devices (90ms/MP on smartphone). We show the efficacy of our method by comprehensive comparative study against existing methods, on real world blur images as well as popular public deblurring data. Quantitative comparison of our work on public benchmark datasets shows significant improvement (+2.19 dB PSNR) over the state of the art while being almost 100 times faster.

Single image haze removal is a challenging and ill-posed problem. The existing haze removal methods in literature, including the recently introduced deep learning methods, model the problem of haze removal as that of estimating intermediate parameters, viz., scene transmission map and atmospheric light. These are used to compute the haze-free image from the hazy input image. Such an approach only focuses on accurate estimation of intermediate parameters, while the aesthetic quality of the haze-free image is unaccounted for in the optimization framework. Thus, errors in the estimation of intermediate parameters often lead to generation of inferior quality haze-free images. In this paper, we present CANDY (Conditional Adversarial Networks based Dehazing of hazY images), a fully end-to-end model which directly generates a clean haze-free image from a hazy input image. CANDY also incorporates the visual quality of haze-free image into the optimization function; thus, generating a superior quality haze-free image. This is one of the first works in literature to propose a fully end-to-end model for single image haze removal. Also, this is the first work to explore the concept of generative adversarial networks for the problem of single image haze removal. CANDY was trained on a synthetically created haze image dataset, while evaluation was performed on challenging synthetic as well as real haze image datasets. The extensive evaluation and comparison results of CANDY reveal that it significantly outperforms existing state-of-the-art haze removal methods in literature, both quantitatively as well as qualitatively.

In low or moderate ambient light conditions, flash image is sharp and less noisy as compared to no-flash image; however, it contains unwanted sharp shadows at silhouettes. In our previous work, we proposed a new flash image shadow detection method which utilizes a pair of flash/no-flash images to detect sharp shadows at silhouettes; however, the method often leads to false shadow detection due its inability to effectively discriminate dark background areas. In this paper, we present an enhanced shadow detection algorithm which shows significant improvement in detection results. A seamless cloning method is used to fill detected shadow regions using no-flash image information. A quantitative as well as qualitative comparison of the proposed shadow detection method on our flash/no-flash image dataset shows that it outperforms current state-of-the-art shadow detection algorithms, both in terms of accuracy as well as speed. The qualitative results of proposed shadow recovery technique show that it generates good quality resultant flash images without sharp shadows.