#THREAD Chandramouleshwara temple🎪is dedicated to bhagwaan Shiva, entrance
Unkal is an area in the Hubli-Dharwad municipality in #Karnataka. It is estimated to be a 900-year-old temple from the Badami Chalukyan era.Chalukyas built this temple in 11th - 12th century @LostTemple7

It is on old Pune – Bangalore highway NH4, about 3 km north of Hubli city centre. The beautiful, natural, scenic Unkal Lake is located in this area.The Chandramouleshwar temple is dedicated to Bhagwan Shiva. It houses two big Shivalingas.
The main shrine has four entrances in all the four directions. The presiding deity of the shrine, Bhagwan Chandramouleshwar faces east. The other Linga is known as Chaturmukha Linga, literally meaning the four faced Linga.
It was built on the lines of Badami, Aihole and Pattadakal temples. Fables say that the Chalukyans wanted to boast their architectural prowess and wealth to their enemies. To do so, the Chalukyans commissioned anonymous artisans to build this temple overnight.
But, the artisans couldn't complete the work within a night and the temple structure remains incomplete. According to some folklore, the sculptor Jaknacharya and his son built this temple.
It is a Monument of National importance and is designated as a Protected Monument under the Ancient Monuments and Archaeological Sites Act (1958) by Archaeological Survey of India (ASI).
Pic 📸Clicked By_ 1st four _Ajay Pmr Way to Shiva & others by Shiv _hiremath

More from Suchi Sharma

More from All

How can we use language supervision to learn better visual representations for robotics?

Introducing Voltron: Language-Driven Representation Learning for Robotics!

Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z

🧵👇(1 / 12)


Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.

Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)

The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (
https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).

The secret is *balance* (3/12)

Starting with a masked autoencoder over frames from these video clips, make a choice:

1) Condition on language and improve our ability to reconstruct the scene.

2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)

By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.

Why is the ability to shape this balance important? (5/12)

You May Also Like