Visão Computacional - Histórico de revisão

V1z: prev years

2018-08-06T12:15:29Z

prev years

← Edição anterior		Edição das 09h15min de 6 de agosto de 2018
Linha 1:		Linha 1:
	This is the homepage for the computer vision course at the undergraduate level being taught in 2018/2 at [http://pt.wikipedia.org/wiki/IPRJ IPRJ]/[http://pt.wikipedia.org/wiki/IPRJ UERJ]. This year will focus on programming 3D technology useful for robotics, motion capture and augmented reality applications such as Kinnect, [https://www.microsoft.com/en-us/hololens Hololens], 3D scanning in Google streetview, and Glass.		This is the homepage for the computer vision course at the undergraduate level being taught in 2018/2 at [http://pt.wikipedia.org/wiki/IPRJ IPRJ]/[http://pt.wikipedia.org/wiki/IPRJ UERJ]. This year will focus on programming 3D technology useful for robotics, motion capture and augmented reality applications such as Kinnect, [https://www.microsoft.com/en-us/hololens Hololens], 3D scanning in Google streetview, and Glass.

			* Links to previous years of the course: [[Visão_Computacional_2013\|2013]]

	[[Imagem:Hologirl.jpeg\|right\|550px]]		[[Imagem:Hologirl.jpeg\|right\|550px]]

V1z: /* Assignment 1 */

2018-08-06T01:02:54Z

Assignment 1

← Edição anterior		Edição das 22h02min de 5 de agosto de 2018
Linha 55:		Linha 55:
	== Homework ==		== Homework ==
	=== Assignment 1 ===		=== Assignment 1 ===
	[[Media:L1-visao-2018.pdf\|Exercise list for Soatto's book, chapter 2]]		* [[Media:L1-visao-2018.pdf\|Exercise list for Soatto's book, chapter 2]]
			* Due date: entregar no inicio da P1

	== Practice Labs ==		== Practice Labs ==

V1z: /* Exams */

2018-08-06T00:59:38Z

Exams

← Edição anterior		Edição das 21h59min de 5 de agosto de 2018
Linha 50:		Linha 50:

	=== Exams ===		=== Exams ===
	P1: 16Aug18 Signature / presence required for course		<b>P1:</b> 16Aug18 Signature / presence required for course
	Final/Sub:		Final/Sub:

	== Homework ==		== Homework ==

V1z: /* Homework */

2018-08-06T00:58:10Z

Homework

← Edição anterior		Edição das 21h58min de 5 de agosto de 2018
Linha 54:		Linha 54:

	== Homework ==		== Homework ==
			=== Assignment 1 ===
			[[Media:L1-visao-2018.pdf\|Exercise list for Soatto's book, chapter 2]]

	== Practice Labs ==		== Practice Labs ==
	=== Download ===		=== Download ===

V1z: /* 3D Reconstruction for Robotics */

2018-08-05T23:49:13Z

3D Reconstruction for Robotics

← Edição anterior		Edição das 20h49min de 5 de agosto de 2018
Linha 135:		Linha 135:
	# Generate your own 3D curve-based reconstruction & visualize it		# Generate your own 3D curve-based reconstruction & visualize it

	~~Evaluation Criteria~~		==== More information ====
	* Whoever completes all steps will score a 10		* Whoever completes all steps will score a 10
	* Bonus: whoever runs on own dataset will have a bonus of 3 points over the maximum project grade. For example, an object rotating in front of a robot.		* Bonus: whoever runs on own dataset will have a bonus of 3 points over the maximum project grade. For example, an object rotating in front of a robot.
			* Due date: 20Ago18 para os que nao forem fazer prova final, ou data da P1 para os que quiserem fazer prova final.

	=== Optional: Toy Piano ===		=== Optional: Toy Piano ===

V1z: /* Part 1 */

2018-08-05T23:38:25Z

Part 1

← Edição anterior		Edição das 20h38min de 5 de agosto de 2018
Linha 119:		Linha 119:
	## Generate and visualize a dense point cloud		## Generate and visualize a dense point cloud
	## Export to meshlab and visualize the cameras and 3D reconstruction		## Export to meshlab and visualize the cameras and 3D reconstruction
	# Bonus ~~points~~: Reconstruct/visualize the dataset using [http://ccwu.me/vsfm/ Visual SfM] or [https://www.blender.org/ Blender]		# +1 Bonus point: generate your own dataset related to robot soccer and perform all steps above
			# +1 Bonus point: Reconstruct/visualize the dataset using [http://ccwu.me/vsfm/ Visual SfM] or [https://www.blender.org/ Blender]

	==== Part 2 ====		==== Part 2 ====

V1z: /* Part 2 */

2018-08-05T23:37:15Z

Part 2

← Edição anterior		Edição das 20h37min de 5 de agosto de 2018
Linha 123:		Linha 123:
	==== Part 2 ====		==== Part 2 ====
	# Build LEMSVXD (copy the code from the professor)		# Build LEMSVXD (copy the code from the professor)
			# Compute curves for each view
	# Visualize epipolar lines across many views		# Visualize epipolar lines across many views
	## Build GUI at `/Users/rfabbri/cprg/vxlprg/lemsvpe/lemsvxl/contrib/rfabbri/mw/app`		## Build GUI at `/Users/rfabbri/cprg/vxlprg/lemsvpe/lemsvxl/contrib/rfabbri/mw/app`

V1z: /* Part 1 */

2018-08-05T23:36:02Z

Part 1

← Edição anterior		Edição das 20h36min de 5 de agosto de 2018
Linha 115:		Linha 115:
	## Play with some parameters		## Play with some parameters
	### Reconstruction -> Automatic Reconstruction -> Dense model (needs CUDA)		### Reconstruction -> Automatic Reconstruction -> Dense model (needs CUDA)
	## With [http://www.regard3d.org/ Regard3D]		## With [http://www.regard3d.org/ Regard3D] (I've obtained nicer results more easily with Regard3D)
			### Follow [http://www.regard3d.org/index.php/documentation/tutorial the tutorial]
			## Generate and visualize a dense point cloud
			## Export to meshlab and visualize the cameras and 3D reconstruction
	# Bonus points: Reconstruct/visualize the dataset using [http://ccwu.me/vsfm/ Visual SfM] or [https://www.blender.org/ Blender]		# Bonus points: Reconstruct/visualize the dataset using [http://ccwu.me/vsfm/ Visual SfM] or [https://www.blender.org/ Blender]

V1z: /* Part 1 */ more instructions on part 1

2018-08-05T23:29:57Z

Part 1: more instructions on part 1

← Edição anterior		Edição das 20h29min de 5 de agosto de 2018
Linha 104:		Linha 104:
	==== Part 1 ====		==== Part 1 ====
	# To practice, [https://drive.google.com/open?id=1xfeO1wVeDUs0NoSsw8TT5Gwnp3cLxHo9 download the "Kermit the Frog" test dataset here] (copied from [http://www.cs.cornell.edu/~snavely/bundler/ Bundler]).		# To practice, [https://drive.google.com/open?id=1xfeO1wVeDUs0NoSsw8TT5Gwnp3cLxHo9 download the "Kermit the Frog" test dataset here] (copied from [http://www.cs.cornell.edu/~snavely/bundler/ Bundler]).
	# Download [https://colmap.github.io COLMAP] 3D reconstruction and structure from motion software (use Linux)		# Download [https://colmap.github.io COLMAP] 3D reconstruction and structure from motion software (use Linux), and also [Regard3D]
	# Run a 3D reconstruction and visualize it.		# Run a 3D reconstruction and visualize it.
	# Take some screenshots and put them on your report.		## With COLMAP
			### Go to Reconstruction -> Automatic Reconstruction
	~~Visualize~~ the dataset using [http://ccwu.me/vsfm/ Visual SfM] or [https://www.blender.org/ Blender]		### Set the image folder as `kermit` (where the `.jpg` images are stored)
			### Set the workspace folder as any folder (I create a folder `kermit-work` for that)
			### Click "Run"
			### Visualize the reconstruction. Go to Render -> Render options, and increase the point size
			## Take some screenshots and put them on your report
			## Play with some parameters
			### Reconstruction -> Automatic Reconstruction -> Dense model (needs CUDA)
			## With [http://www.regard3d.org/ Regard3D]
			# Bonus points: Reconstruct/visualize the dataset using [http://ccwu.me/vsfm/ Visual SfM] or [https://www.blender.org/ Blender]

	==== Part 2 ====		==== Part 2 ====

V1z: /* 3D Reconstruction for Robotics */ updated instructions

2018-08-05T20:31:24Z

3D Reconstruction for Robotics: updated instructions

← Edição anterior		Edição das 17h31min de 5 de agosto de 2018
Linha 102:		Linha 102:


	~~Steps:~~		==== Part 1 ====
	# To practice, download ~~this~~ dataset ~~containing multiple images, with reconstructed 3D points and camera models~~: https://github.~~com/rfabbri/pavilion-multiview-3d-dataset~~		# To practice, [https://drive.google.com/open?id=1xfeO1wVeDUs0NoSsw8TT5Gwnp3cLxHo9 download the "Kermit the Frog" test dataset here] (copied from [http://www.cs.cornell.edu/~snavely/bundler/ Bundler]).
	#~~# This takes~~ a ~~long time to download~~. ~~First, practice with the capitol-right-dome dataset sent by the professor~~.		# Download [https://colmap.github.io COLMAP] 3D reconstruction and structure from motion software (use Linux)
	# Visualize the dataset using [http://ccwu.me/vsfm/ Visual SfM] or [https://www.blender.org/ Blender]		# Run a 3D reconstruction and visualize it.
			# Take some screenshots and put them on your report.

			Visualize the dataset using [http://ccwu.me/vsfm/ Visual SfM] or [https://www.blender.org/ Blender]

			==== Part 2 ====
	# Build LEMSVXD (copy the code from the professor)		# Build LEMSVXD (copy the code from the professor)
	# Visualize epipolar lines across many views		# Visualize epipolar lines across many views