Meta tags:
description=Personal website homepage of Mukul Khanna. ;

Headings (most frequently used words):

mukul, khanna, news, select, publications,

Text of the page (most frequently used words):
and (42), the (25), mukul (12), #khanna (12), 2024 (11), underwater (10), for (10), network (8), dhruv (8), image (7), website (7), from (7), batra (7), feedback (6), vision (6), code (6), learning (6), reconstruction (6), arxiv (6), goat (6), using (5), hdr (5), single (5), ldr (5), with (5), normal (5), chang (5), prof (5), vehicles (4), systems (4), dynamic (4), navigation (4), high (4), forward (4), neural (4), over (4), scenes (4), cvpr (4), ursim (3), open (3), simulation (3), while (3), robots (3), abs (3), range (3), been (3), deep (3), feed (3), have (3), are (3), state (3), multiple (3), fhdr (3), our (3), raman (3), estimation (3), often (3), maps (3), outdoor (3), performance (3), conference (3), roozbeh (3), mottaghi (3), devendra (3), singh (3), chaplot (3), angel (3), manolis (3), savva (3), theophile (3), gervet (3), sriram (3), yenamandra (3), hssd (3), bench (3), podcast (3), research (3), also (3), advised (3), 2025 (2), presents (2), source (2), unmanned (2), robotics (2), ros (2), real (2), unity3d (2), like (2), these (2), complex (2), environment (2), provides (2), simulate (2), control (2), realistic (2), architecture (2), manipulation (2), 2019 (2), generation (2), exposure (2), low (2), has (2), various (2), networks (2), cnns (2), better (2), where (2), level (2), features (2), conventional (2), this (2), parameters (2), end (2), art (2), shanmuganathan (2), zeeshan (2), khan (2), computer (2), robot (2), most (2), approaches (2), indoor (2), they (2), exhibit (2), significant (2), building (2), facades (2), more (2), adversarial (2), that (2), synthetic (2), dataset (2), madhava (2), krishna (2), 2022 (2), deephs (2), devi (2), parikh (2), chris (2), paxton (2), zsolt (2), kira (2), clegg (2), rss (2), shah (2), matthew (2), any (2), thing (2), publications (2), joined (2), apple (2), products (2), group (2), spring (2), summer (2), intern (2), jan (2), accepted (2), georgia (2), tech (2), waymo (2), engineer (2), about (2), had (2), privilege (2), was (2), working (2), copyright, paper, framework, uuvs, developed, operating, system, time, game, engine, called, enable, implement, test, study, analyze, minimizing, cost, disruption, user, intuitive, way, capable, simulating, model, mission, planning, support, sensor, modules, physics, collision, kinematics, highly, configurable, software, adaptive, algorithms, processing, technology, harshit, nagar, panaiyappan, pushkal, katara, simulator, made, possible, due, recent, advances, convolutional, proposed, representations, utilize, power, exploit, idea, initial, guided, hidden, recurrent, unlike, pass, learned, iterations, enables, create, coarse, fine, representation, leading, improved, every, iteration, advantages, standard, include, early, ability, quality, fewer, design, dense, block, propose, qualitative, quantitative, evaluations, show, superiority, approach, methods, globalsip, surface, essential, component, several, pipelines, problem, extensively, studied, geared, towards, rely, modalities, depth, views, accurate, pose, greater, challenge, lighting, variation, contain, occluders, structures, ridden, numerous, windows, protrusions, supervised, schemes, excel, but, not, competitive, when, trained, deployed, environments, furthermore, involve, architectures, require, many, trainable, tackle, challenges, present, scheme, regularizes, output, appear, small, number, precisely, annotated, examples, method, lightweight, simpler, improving, least, across, metrics, evaluate, against, map, observe, enhancements, 2021, crv, tanu, sharma, ayyappa, swamy, thatavarthy, view, images, bf2normalnet, international, pattern, recognition, icpr, parth, shettiwar, hdrvideo, speed, video, hdrv, pdf, ruta, desai, vincent, cartillier, sameer, dharur, samyak, datta, episodic, memory, question, answering, emqa, 2023, corl, neurips, competition, track, yonatan, bisk, john, turner, alex, william, vidhi, jain, tsung, yen, yang, austin, wang, karmesh, yadav, arun, ramachandran, homerobot, vocab, mobile, ovmm, kavit, jitendra, malik, saurabh, gupta, yeon, min, eric, undersander, alexander, brennan, shacklett, sanjay, haresh, hanxiao, jiang, yongsen, mao, habitat, 200, analysis, scene, scale, realism, tradeoffs, objectgoal, ram, ramrakhya, gunjan, chhablani, benchmark, multi, modal, lifelong, select, papers, feb, grateful, awarded, gatech, college, computing, marshall, williamson, fellowship, mar, apr, graduated, dec, perception, org, news, side, dabbled, hosting, interviewed, academics, scientists, their, insights, journeys, likewise, managed, hosted, humans, notebooks, distant, past, interning, center, iiit, hyderabad, iit, gandhinagar, shanmuga, previously, student, embodied, before, closely, simon, fraser, university, experience, current, toggle,

Text of the page (random words):
mukul khanna toggle navigation about current experience podcast publications mukul khanna i m an ml engineer at waymo previously i was an ms student at georgia tech advised by prof dhruv batra working on computer vision and embodied ai research i was also an ml intern with apple s vision products group over spring and summer of 2024 and before that have had the privilege of working closely with prof manolis savva and prof angel chang at simon fraser university in a distant past i have also had the privilege of interning at the robotics research center at iiit hyderabad advised by prof k madhava krishna and at iit gandhinagar advised by prof shanmuga raman on the side i dabbled with hosting the deep neural notebooks podcast where i interviewed academics and scientists about their research insights and journeys likewise i also managed the humans of ai podcast hosted by devi parikh and dhruv batra news jan 2025 joined waymo as an ml engineer in the perception org dec 2024 graduated from georgia tech apr 2024 goat go to any thing accepted at rss 2024 mar 2024 grateful to be awarded the marshall d williamson fellowship by gatech college of computing feb 2024 2 papers hssd and goat bench accepted at cvpr 2024 jan 2024 joined apple vision products group as a spring and summer intern select publications goat bench goat bench a benchmark for multi modal lifelong navigation mukul khanna ram ramrakhya gunjan chhablani sriram yenamandra theophile gervet matthew chang devendra singh chaplot zsolt kira dhruv batra and roozbeh mottaghi cvpr 2024 arxiv website code hssd habitat synthetic scenes dataset hssd 200 an analysis of 3d scene scale and realism tradeoffs for objectgoal navigation ️ mukul khanna yongsen mao hanxiao jiang sanjay haresh brennan shacklett dhruv batra alexander clegg eric undersander angel x chang and manolis savva cvpr 2024 arxiv website code goat goat go to any thing matthew chang theophile gervet mukul khanna sriram yenamandra dhruv shah so yeon min kavit shah chris paxton saurabh gupta dhruv batra roozbeh mottaghi jitendra malik and devendra singh chaplot rss 2024 arxiv website ovmm homerobot open vocab mobile manipulation sriram yenamandra arun ramachandran karmesh yadav austin wang mukul khanna theophile gervet tsung yen yang vidhi jain alex william clegg john turner zsolt kira manolis savva angel chang devendra singh chaplot dhruv batra roozbeh mottaghi yonatan bisk and chris paxton conference on robot learning corl neurips competition track 2023 arxiv website code emqa episodic memory question answering ️ samyak datta sameer dharur vincent cartillier ruta desai mukul khanna dhruv batra and devi parikh cvpr 2022 website pdf deephs hdrv deephs hdrvideo deep high speed high dynamic range video reconstruction zeeshan khan parth shettiwar mukul khanna and shanmuganathan raman international conference on pattern recognition icpr 2022 arxiv bf2normalnet building facades to normal maps adversarial learning from single view images mukul khanna tanu sharma ayyappa swamy thatavarthy and k madhava krishna conference on robots and vision crv 2021 abs website code surface normal estimation is an essential component of several computer and robot vision pipelines while this problem has been extensively studied most approaches are geared towards indoor scenes and often rely on multiple modalities depth multiple views for accurate estimation of normal maps outdoor scenes pose a greater challenge as they exhibit significant lighting variation often contain occluders and structures like building facades are often ridden with numerous windows and protrusions conventional supervised learning schemes excel in indoor scenes but do not exhibit competitive performance when trained and deployed in outdoor environments furthermore they involve complex network architectures and require many more trainable parameters to tackle these challenges we present an adversarial learning scheme that regularizes the output normal maps from a neural network to appear more realistic by using a small number of precisely annotated examples our method presents a lightweight and simpler architecture while improving performance by at least 1 5x across most metrics we evaluate our approaches against the state of the art on normal map estimation on a synthetic and a real outdoor dataset and observe significant performance enhancements fhdr fhdr hdr image reconstruction from a single ldr image using feedback network zeeshan khan mukul khanna and shanmuganathan raman globalsip 2019 abs arxiv code high dynamic range hdr image generation from a single exposure low dynamic range ldr image has been made possible due to the recent advances in deep learning various feed forward convolutional neural networks cnns have been proposed for learning ldr to hdr representations to better utilize the power of cnns we exploit the idea of feedback where the initial low level features are guided by the high level features using a hidden state of a recurrent neural network unlike a single forward pass in a conventional feed forward network the reconstruction from ldr to hdr in a feedback network is learned over multiple iterations this enables us to create a coarse to fine representation leading to an improved reconstruction at every iteration various advantages over standard feed forward networks include early reconstruction ability and better reconstruction quality with fewer network parameters we design a dense feedback block and propose an end to end feedback network fhdr for hdr image generation from a single exposure ldr image qualitative and quantitative evaluations show the superiority of our approach over the state of the art methods ursim open source simulator for unmanned underwater vehicles using ros and unity3d pushkal katara mukul khanna harshit nagar and a panaiyappan underwater technology ut 2019 abs website code the paper presents ursim an open source 3d underwater simulation framework for unmanned underwater vehicles uuvs developed using robotics operating system ros and a real time game engine called unity3d simulation systems like these enable to implement test study and analyze complex systems while minimizing cost and disruption to the environment ursim provides the user an intuitive way to simulate underwater vehicles and robots it is capable of simulating feedback control systems dynamic model underwater vision and mission planning for underwater vehicles and robots the simulation provides support for underwater sensor modules underwater physics collision kinematics and is highly configurable to simulate a realistic underwater environment the software architecture is adaptive to algorithms for control systems image processing navigation and manipulation copyright 2025 mukul khanna